r graphics ii: graphics for exploratory data...

72
Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links UCLA Department of Statistics Statistical Consulting Center R Graphics II: Graphics for Exploratory Data Analysis Irina Kukuyeva [email protected] April 26, 2010 Irina Kukuyeva [email protected] R Graphics II: Graphics for Exploratory Data Analysis UCLA SCC

Upload: vukhanh

Post on 10-Apr-2018

230 views

Category:

Documents


5 download

TRANSCRIPT

Page 1: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

UCLA Department of StatisticsStatistical Consulting Center

R Graphics II Graphics for Exploratory Data Analysis

Irina Kukuyevaikukuyevastatuclaedu

April 26 2010

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Outline

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary PlotsBasic PlotsLooking at DistributionsIdentify-ing ObservationsExercise I

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot I

1 Step 1 Load the Data

2 library(alr3)

3 data(UN2)

4 attach(UN2)

5 Step 2 Subset appropriately

6 ind lt-which(Purban gt50)

7 Step 3 Plot

8 plot(logFertility~logPPgdp xlab=logPPgdp

ylab=logFertility main=logGDP vs

logFertility Plot)

9 points(logFertility[ind]~logPPgdp[ind] col=

red pch =19)

10 legend(topright pch=c(119) col=12 c(

Purban lt50 Purban gt=50))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot II

8 10 12 14

00

05

10

15

20

logGDP vs logFertility Plot

logPPgdp

logF

ertil

ity

Purbanlt50Purbangt=50

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 2: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Outline

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary PlotsBasic PlotsLooking at DistributionsIdentify-ing ObservationsExercise I

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot I

1 Step 1 Load the Data

2 library(alr3)

3 data(UN2)

4 attach(UN2)

5 Step 2 Subset appropriately

6 ind lt-which(Purban gt50)

7 Step 3 Plot

8 plot(logFertility~logPPgdp xlab=logPPgdp

ylab=logFertility main=logGDP vs

logFertility Plot)

9 points(logFertility[ind]~logPPgdp[ind] col=

red pch =19)

10 legend(topright pch=c(119) col=12 c(

Purban lt50 Purban gt=50))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot II

8 10 12 14

00

05

10

15

20

logGDP vs logFertility Plot

logPPgdp

logF

ertil

ity

Purbanlt50Purbangt=50

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 3: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary PlotsBasic PlotsLooking at DistributionsIdentify-ing ObservationsExercise I

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot I

1 Step 1 Load the Data

2 library(alr3)

3 data(UN2)

4 attach(UN2)

5 Step 2 Subset appropriately

6 ind lt-which(Purban gt50)

7 Step 3 Plot

8 plot(logFertility~logPPgdp xlab=logPPgdp

ylab=logFertility main=logGDP vs

logFertility Plot)

9 points(logFertility[ind]~logPPgdp[ind] col=

red pch =19)

10 legend(topright pch=c(119) col=12 c(

Purban lt50 Purban gt=50))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot II

8 10 12 14

00

05

10

15

20

logGDP vs logFertility Plot

logPPgdp

logF

ertil

ity

Purbanlt50Purbangt=50

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 4: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot I

1 Step 1 Load the Data

2 library(alr3)

3 data(UN2)

4 attach(UN2)

5 Step 2 Subset appropriately

6 ind lt-which(Purban gt50)

7 Step 3 Plot

8 plot(logFertility~logPPgdp xlab=logPPgdp

ylab=logFertility main=logGDP vs

logFertility Plot)

9 points(logFertility[ind]~logPPgdp[ind] col=

red pch =19)

10 legend(topright pch=c(119) col=12 c(

Purban lt50 Purban gt=50))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot II

8 10 12 14

00

05

10

15

20

logGDP vs logFertility Plot

logPPgdp

logF

ertil

ity

Purbanlt50Purbangt=50

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 5: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Basic Plot II

8 10 12 14

00

05

10

15

20

logGDP vs logFertility Plot

logPPgdp

logF

ertil

ity

Purbanlt50Purbangt=50

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 6: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts I

Displays two categorical variables at a time 1

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 barplot(table(gender hand) col = c(skyblue

blue) main = Segmented Bar Plot n

of Gender)

4 legend(topleft c(femalesmales) col =

c(skyblue blue) pch = 16 inset =

005)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 7: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Segmented bar charts II

ambidextrous left right

Segmented Bar Plot

of Gender

0200400600800

females

males

ambidextrous left right

female 9 67 806male 11 45 387

1These two slides are modified from the SCC Mini-Course rdquoIntroductoryStatistics with Rrdquo by Mine Cetinkaya

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 8: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts I

To compare values for variables in each category

1 Step 1 Load the data

2 data(iris)

3 attach(iris)

4 Step 2 Calculate means for each species

5 aggregate(iris[ -5] list(Species = Species)

mean)-gta

6 Step 3 Assign row names

7 rownames(a)lt-a[ 1]

8 Step 4 Plot

9 dotchart(t(a[ -1]) xlim = c(010) main=

Plots of Means for Iris Data Set xlab=

Mean Value)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 9: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Basic Plots

Dot charts II

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

SepalLength

SepalWidth

PetalLength

PetalWidth

setosa

versicolor

virginica

0 2 4 6 8 10

Plots of Means for Iris Data Set

Mean Value

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 10: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

HistogramsAdding Summary Statistics to Plots

Add the mean and median to a histogram

1 hist(ageinmonths main=

Histogram of Age (Mo))

2 abline(v=mean(ageinmonths)

col = blue lwd=3)

3 abline(v=median(ageinmonths

) col = red lwd=3)

4 legend(topright c(Mean

Median) pch = 16

col = c(blue red))

Histogram of Age (Mo)

ageinmonths

Fre

quen

cy200 250 300 350

010

020

030

040

0

MeanMedian

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 11: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IChecking Normality

One of the methods to test for normality of a variable is to look atthe histogram (the sample density is in red the theoretical normaldensity in blue)

1 data(presidents)

2 hist(presidents prob=T ylim=c(0 004)

breaks =20)

3 lines(density(presidents narm=TRUE) col=

red)

4 mult-mean(presidents narm=TRUE)

5 sigma lt-sd(presidents narm=TRUE)

6 xlt-seq(10100 length =100)

7 ylt-dnorm(xmu sigma)

8 lines(xylwd=2col=blue)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 12: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Histograms IIChecking Normality

Histogram of Presidents Approval Ratings

presidents

Den

sity

20 30 40 50 60 70 80 90

000

001

002

003

004

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 13: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot I

Another method of looking at the distribution of the data is viaboxplot

1 data(quakes)

2 Subset the magnitude

3 ind lt-ifelse(quakes[ 4]lt45 0 1)

4 ind lt-asfactor(ind)

5 library(lattice)

6 bwplot(quakes[ 4]~ind xlab=c(Mag lt45 Mag

gt=45) ylab=Magnitude)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 14: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Box and Whisker Plot IIFiji EQ since 1964

Maglt45 Maggt=45

Mag

nitu

de

40

45

50

55

60

65

0 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 15: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot I

An alternative to the boxplot is the beanplot()

1 library(beanplot)

2 par(mfrow=c(12))

3 data(airquality)

4 boxplot(airquality[ 2] main=Boxplot xlab=

Solar)

5 beanplot(airquality[ 2] main=Beanplot

xlab=Solar)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 16: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Beanplot II

050

100

150

200

250

300

Boxplot

Solar

010

020

030

040

0

Beanplot

SolarIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 17: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots I

A method of looking at the distribution and correlation of the datais via scatterplotmatrix()

1 data(quakes)

2 library(car)

3 scatterplotmatrix(quakes[ 14])

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 18: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Scatterplots II

||| ||| ||| || |||| | ||| | | ||| | | ||| || ||| || | |||| || ||| | || |||| | |||| || || | || ||| | ||| || || | |||| | || |||| ||| || | ||| | || || || | || ||| |||||| |||| ||| || |||| | |||| | |||| | || || || ||| || || ||| || | ||| ||| | | || | || | ||| | || ||| ||| || || || |||| ||| | || | || ||| || || || || || || | ||| || || || | || ||||| | || || || ||| || | ||||| | |||| || | ||| |||| ||| || ||| | || | |||| || || | ||| ||| || | || || || |||| ||||| |||| ||| || || | || |||| ||| |||||||||||| ||| ||| | || || || ||| || | | ||| | ||| || |||| || ||| || || || | ||| | || || || | ||| ||| || || | |||| ||| |||| || | ||| || || ||| | || |||| ||||| | || | |||| | ||||| | |||| | | | |||| | | || ||| ||| || || || ||| || | | ||| || | ||||| | || | | || || | || || | | ||| ||| || || ||| |||||| || |||| |||||| | |||| || || ||| || || | ||| || || ||| ||| || || | || | |||||| || || |||| | | || ||| | | ||| || ||| | | |||| | | || ||| | || |||| || | || || || || ||| | || ||| || || ||| || ||| | ||| | ||| || |||||| | | | || |||| | || ||| ||||| ||| | || ||| || ||| | ||| | |||| || | ||| ||| ||| ||||| ||| |||| || | ||| ||| ||||||| ||||| | || || || || || || ||| |||| || || ||| | | || || ||||| ||| || ||||| |||| ||| | | | || || ||| || | ||| || ||| ||| || ||| || || ||| ||||| ||| ||| | |||| || || || || ||| || | ||| | || | | || || ||| | || ||| | || | |||| || | || | ||| |||| || || || ||||| |||| | ||| || | ||| || || ||| || |||| | ||| | || || ||| || ||

lat

165 170 175 180 185

40 45 50 55 60

minus35

minus25

minus15

165

170

175

180

185

|| ||| || ||| || ||| || ||| || | || || | | ||| || ||| ||| | || || ||| | || || | |||| || | ||| | ||| | |||| | || || ||| | | || || | |||| || ||| || | || | || ||| || ||||||||| || | | ||| || | | ||| ||| ||| || || || | || | | || || | || ||| ||| ||| | ||| ||| |||| || || | ||| |||| ||| ||| | |||| || ||| || || || || || | || || || |||| | ||| || | ||| ||| | ||| | ||| |||| | || || || || ||| | | || | ||||| || ||| | |||| ||| | || ||| || ||| |||| | ||| | || |||| || |||| ||| |||| ||| || || || | ||||||| |||||| |||| | |||| ||| | |||| || |||| ||| | || ||||| || ||||| | |||||| | ||| || || || || | | ||| | ||| || || | || | || || | || || ||| || || || | || |||| ||| | || | ||| ||||| || ||| | ||| ||| || || |||| || ||| || | | || ||| | |||| | |||| | || || | || | || || || |||| |||| || || ||| |||||||| | || ||| ||||||| | || || ||| ||| || || ||| | || || || || | || | ||| ||| || |||||| | |||| ||| | || || || ||| || | ||| || ||| || || || || | || || || |||| || | | |||| ||| ||| || ||| | || | ||| || || | || | |||| ||||| | || || | || | || ||| || || || | | ||| |||| | ||| |||||| ||| ||| ||| ||| ||| | ||| || ||| ||| |||| ||| ||| |||| ||| ||| | |||| | |||| ||||| | || || || | ||||| | || ||| | |||| || || || |||| |||| || | || | ||| | |||| | ||| || || ||| || ||| | | ||| ||| || |||| | |||| |||| ||| || ||| || ||| || |||||| ||| ||| | ||||| |||||| ||| ||| || || |||| || ||||| |||| |||| | || | ||| |||||| || ||| | ||||| || || || || | ||| || ||| | ||

long

| || | ||| || ||| | || || || ||| || | || | || || | || || ||| || || || || || ||| || || ||| || || || | ||| | || | ||| | ||| | || ||| ||| | | ||| |||| || | || ||| || | | | | ||||| | ||| |||| | || || || | ||| || | |||| |||| | || | || | || ||| |||||| | | || || || || || || || | | || | ||||| | || ||| | || ||| || ||| ||||| | || || || || | || || || || || | || || |||| | | | || || || || | ||| ||| || ||| || | || ||| |||| ||| || || | || || || |||| | || | ||| | ||| ||| ||| || ||| ||| | |||| || ||| || | |||||| ||||| ||||||| ||| || || | || |||| || || || | ||| || || | || | || |||| || || | ||| | || || || ||| ||||| || || | || ||| || | ||| || | |||| | | ||||| ||| | | ||| || ||| || | || |||| || | || ||| || ||| || |||| || | | | || | |||||| |||| || | |||| |||| | || | || || || || || | |||| ||| | ||| || |||| || || |||| || ||| || || | | ||| ||| || || || ||||| || | ||| | || ||||| || ||||| |||||| |||| | || ||| ||| | ||| || || | ||| || || || | ||| | |||| | | ||| | || || ||| | || |||| || || ||||| | |||| | || | || || || | || | ||| || | |||| || | || || | | ||| | || | ||| | || || ||| ||| | | || ||| | || || || || || ||||| | | ||||| ||||| || || || || || |||| || || | | ||| || | || || || ||| | || | ||| | |||| |||| ||| |||| | || | ||| | || | || || | ||| |||||| || ||| | ||||| | ||| | ||| || || | ||| || | | |||| | | || | ||| || |||| || ||| || || | ||| || | ||| ||| ||| || | ||| || || || || |||| |||| || | ||||| |||| | || || ||| ||| || | || || || ||| ||| || ||| || |||| |

depth

100

300

500

700

minus35 minus25 minus15

40

45

50

55

60

100 300 500 700

|| |||| || || | ||| || |||| || | | || | ||| | | || || | || | |||| | |||| ||| ||| | || || || || ||| || || || |||| || ||| ||| | ||| | || |||| | |||| | |||| | | ||||| ||| || || || ||| || ||| || | || | || || || |||| | | || | ||| ||| | ||| ||| | | ||||| |||| |||| ||| ||| || ||| |||| || ||| ||| |||| || || ||| |||| ||| || || | || | |||| | ||| | || ||| || ||| | || || |||| | || | ||||| ||| ||| || | || || | ||| || | || || ||||| || ||| || |||| || || | | | || ||| || || ||| ||| || |||| | ||| || || ||||| || ||| || || | | || ||| || | | |||| | |||| || || ||| || ||| || || ||| || || ||||| |||| ||| | ||| | |||| || || ||| ||||| || ||| | || || | || ||| | |||| ||| ||| ||| | || ||| || |||| || | |||| ||| || ||| ||| |||| || || | |||| || || | || |||| ||| | ||| || || | || | ||| | |||| || || || || || ||| ||| ||| ||| || | ||| | || |||| || | || |||| ||| ||| | || | | || | || || ||| | ||| |||| ||| || | || | || |||| || ||| || || || || ||| ||| ||| || || || || |||| | | ||| ||| ||| ||| ||| || | || || ||| | || | || ||| || | || |||| | | ||| | || ||| || || || | |||| | || ||| | | ||| || | || | | || ||| | | |||| |||| | || | ||| | || || ||| | |||| | | || || ||| || ||| | || ||| |||| || ||| || ||| || | ||| ||| | || ||| ||| || || | | |||| | ||| || | || | |||| || ||| || | |||| | || | ||| || | |||| || || | || ||| ||| | || || | |||||| || ||||| | || || ||||| || | || ||| || ||| || || || |||| |||| |||| | | || || ||| || || | || || ||| ||| | || || || ||| ||| || | | ||| |

mag

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 19: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 1

A method of checking equality of distributions is via qq()

1 library(lattice)

2 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

3 attach(survey)

4 qq(gender~ageinmonths)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 20: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 1

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 21: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IApproach 2

Another method for checking equality of distributions is viabeanplot()

1 survey = readtable(httpwwwstatuclaedu

~minestudents_survey_2008 txt header =

TRUE sep = t)

2 attach(survey)

3 beanplot(ageinmonths ~ gender data=survey

col=list(grey (05) c(grey (08)white))

border = NA overallline = median ll

=001 side=both)

4 legend(bottomleftfill=c(grey (05) grey (08)

) legend=c(FemaleMale))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 22: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Looking at Distributions

Checking Equality of Distributions IIApproach 2

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 23: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IPreliminaries

1 Generate data and fit a regression curve

2 setseed (3012008)

3 x=rnorm (100) y=-x+I(x^2) +rnorm (100)

4 fit lt-lm(y~x+I(x^2)) fit

Intercept x x2

01307 -09701 09549

1 Plot the resulting regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 24: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations IIPreliminaries

minus2 minus1 0 1 2

minus2

02

46

x

y

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 25: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations I

Left-click on the observations in the graphics window to see therow numberRight-click on the observation to exit the function

1 Plot the data and fit the regression curve

2 plot(y~x pch =19)

3 curve(expr=fit [[1]][1]+ fit [[1]][2]x+fit

[[1]][3]I(x^2) from=range(x)[1] to=

range(x)[2] add=TRUE col=blue lwd=2)

4 Identify the outlying observations

5 index lt-identify(y~x) index

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 26: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Identify-ing Observations

Identify-ing Observations II

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 27: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise I

Exercise I

Compare the distributions of the sepal widths for the three speciesof irises (using the iris data set)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 28: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series PlotsMultivariate PlotsExercise II

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 29: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 1

To plot more than variables one at a time use plot()

1 Convert data to a time series via ts() or

zoo()

2 data(airquality)

3 alt-airquality[ 13]

4 time lt-ts(1 nrow(a) start=c(1973 5)

frequency =365)

5 If your data is stored as a data frame

6 coerce it to be a matrix via asmatrix ()

7 class(a)

8 amat lt-asmatrix(a)

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 30: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 1

11 Make a time series of the two (or more

variables)

12 library(zoo)

13 namezoo lt-zoo(cbind(amat[ 1] amat[ 2]))

14 colnames(namezoo)lt-c(Ozone Solar)

15 Plot the variables

16 plot(namezoo)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 31: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIIApproach 1

Ozo

ne

050

100

150

0 50 100 150

010

020

030

0

Sol

ar

Time

Plots of Ozone and Solar

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 32: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

To plot more than variables one at a time use mvtsplot()2

1 Load the R Code for the function

2 source(httpwwwbiostatjhsphedu~rpengRR

mvtsplotmvtsplotR)

3 After processing data as in Approach 1

4 Plot the variables

5 mvtsplot(namezoo)

6 Purple=low grey=medium green=high white=

missing values

2For documentation go to wwwjstatsoftorgv25c01paperIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 33: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series PlotsApproach 2

Ozone

Solar

0

50

100

150

200

250

300

Leve

l

0 20 40 60 80 100 120 140

0 50 100 150 200 250 300

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 34: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IApproach 3

To plot more than variables one at a time use xyplot()

1 After processing data as in Approach 1

2 load both libraries

3 library(lattice)

4 library(zoo)

5 data(EuStockMarkets)

6 zlt-EuStockMarkets

7 xyplot(z screen = c(1122) col = 14

strip = FALSE key = list(title=

EuStockMarkets columns=2 points=FALSE

lines=TRUE col=14 text=list(colnames(z)

)))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 35: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Multivariate Plots

Multivariate Time Series Plots IIApproach 3

Index

2000

4000

6000

8000

1992 1994 1996 1998

2000

3000

4000

5000

6000

EuStockMarketsDAXSMI

CACFTSE

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 36: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise II

Exercise II

Analyze the EuStockMarkets data using the functionmtvsplot() What stands out and why

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 37: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical PlotsMapsProjection MapsExercise III

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 38: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic MapsMap of Fiji Earthquakes Since 1964

To overlay a map to a plot containing latitude and longitude loadthe package maps

1 data(quakes)

2 library(maps)

3 plot(quakes[ 2]

quakes[ 1] xlim=

c(100 190) ylim=

c(-40 0))

4 map(world add=T)

100 120 140 160 180

minus40

minus30

minus20

minus10

0

quakes[ 2]

quak

es[

1]

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 39: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IMap of Fiji Earthquakes Since 1964

To include information on magnitude of earthquake use cex

argument of the plot() function

1 Step 1 Recode magnitude to have 3

categories only

2 ind lt-which(mag lt5)

3 ind2 lt-which(mag lt6 amp mag gt=5)

4 ind3 lt-which(mag gt=6)

5 color lt-rep(NA length(mag))

6 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

7

8

9

10

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 40: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIMap of Fiji Earthquakes Since 1964

11 Step 2 Plot

12 plot(quakes[ 2] quakes[ 1] cex=color pch

=19 col=color xlim=c(100 190) ylim=

c(-400) xlab=Longitude ylab=Latitude

)

13 legend(topright pch=19 col=13 c(Mag lt5

5lt=Mag lt6 Mag gt6) ptcex =13)

14 map(world add=T)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 41: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Maps

Geographic Maps IIIMap of Fiji Earthquakes Since 1964

100 120 140 160 180

minus40

minus30

minus20

minus10

0

Longitude

Latit

ude

Maglt55lt=Maglt6Maggt6

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 42: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IMap of Fiji Earthquakes Since 1964

For a different perspective of a map use mapproject()

1 library(mapproj)

2 library(maps)

3 m lt- map(rsquoworldrsquoplot=FALSE)

4 Projection is Azimuthal with equal -area

5 map(rsquoworldrsquoproj=rsquoazequalarea rsquoorient=c(

longitude =0latitude =180 rotation =0))

6 mapgrid(mcol=2)

7 points(mapproject(list(y=quakes[which(quakes[

4]gt=6) 1]x=quakes[which(quakes[ 4] gt=6)

2]))col=bluepch=xcex=2)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 43: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Projection Maps IIMap of Fiji Earthquakes Since 1964

minus180 minus150minus100

minus50

0

50

100150 180

minus85

minus60

minus40

minus20

0

20

40

60

84

xxxxx

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 44: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Projection Maps

Bonus Feature of the maps package

To determine in which part of the world the observations are(based on latitude and longitude) use mapwhere()

1 inwhatcountry lt-mapwhere(database=world

quakes[ 2] quakes[ 1])

To determine which observations are in the ocean

1 Number of points in ocean after filtering

2 ind lt-sum(isna(inwhatcountry)) ind

3 Number of observations 1000

4 Number in Ocean 993

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 45: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise III

Exercise III

For the ozone data set 3 determine the region of the world thatthe observations correspond to and overlay the appropriate map

3httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 46: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plotslattice libraryggplot2 libraryrgl librarySaving Plots as a PDFExercise IV

5 Simulation Plots

6 Useful Links for RIrina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 47: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 1 Using wireframe()

1 library(lattice)

2 wireframe(volcano

colregions =

terrain

colors (100) asp =

1 colorkey=TRUE

drape=TRUE

scales = list(

arrows = FALSE))20

40

60

80

10

20

30

4050

60

100

120

140

160

180

rowcolumn

volcano

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 48: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 2 Same image with the levelplot() function

1 library(lattice)

2 levelplot(volcano

asp=1 colregions

=terraincolors)

row

colu

mn

10

20

30

40

50

60

20 40 60 80

100

120

140

160

180

200

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 49: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

lattice library

3D Images with Package lattice

Method 3 Same image with the image() function

1 xlt-10(1 nrow(volcano)

)

2 ylt-10(1 ncol(volcano)

)

3 image(x y volcano

col = terrain

colors (100) axes =

TRUE)

4 contour(x y volcano

add = TRUE col =

1)200 400 600 800

100

200

300

400

500

600

x

y

100

100

100

110

110

110

110

120

130

140

150

160

160

170

170

180

180

190

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 50: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 I

Another way to create 3D images is with the package ggplot2

1 Step 1 Load the data

2 data(quakes)

3 attach(quakes)

4 Step 2 Create a categorical variable for

earthquake magnitude

5 ind lt-which(mag lt5)

6 ind2 lt-which(mag lt6 amp mag gt=5)

7 ind3 lt-which(mag gt=6)

8 color lt-rep(NA length(mag))

9 color[ind]lt-1 color[ind2]lt-2 color[ind3]lt-3

10 color lt-asfactor(color)

11

12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 51: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 II

13 Step 3 Plot

14 library(ggplot2)

15 ggplot(data=quakes aes(long lat))+

16 geom_point(aes(long lat))+

17 borders(world)+

18 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 52: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 III

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 53: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IAdding Another Dimension

To include information on magnitude of earthquake usegeom point()

1 ggplot(data=quakes aes(long lat))+

2 borders(world)+

3 coord_cartesian(xlim=c(100 190) ylim=c(-400)

)+

4 geom_point(data=quakes colour=asnumeric(

color) size=2asnumeric(color))

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 54: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

ggplot2 library

3D Images with Package ggplot2 IIAdding Another Dimension

long

lat

0

100 150

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 55: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

rgl library

3D Images with Package rgl

A way to create 3D interactive images is with the package rgl

1 library(rgl)

2 data(quakes)

3 plot3d(x=quakes[ 2]

y=quakes[ 1] z=

quakes[ 3] xlab=

Longitude ylab=

Latitude zlab=

Depth)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 56: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Static Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 pdf(Volcanopdf width=6 height=6 onefile=

FALSE)

4 wireframe(volcano colregions = terrain

colors (100) asp = 1 colorkey=TRUE

drape=TRUE scales = list(arrows = FALSE))

5 devoff()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 57: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Saving Plots as a PDF

Saving Plots as a PDFFor Dynamic Plots

Note The files will be saved in the folder specified with setwd()

To save a plot in R as a PDF you can use pdf()

1 To save the image to the Desktop

2 setwd(~Desktop)

3 Step 1 Produce the 3D plot

4 plot3d(x=quakes[ 2] y=quakes[ 1] z=quakes

[ 3] xlab=Longitude ylab=Latitude

zlab=Depth)

5 Step 2 Can rotate before taking a snapshot

6 rglsnapshot(quakespng)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 58: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise IV

Exercise IV

Use the rgl library to create a 3D plot of the ozone data set 4

4httpwwwatsuclaedustatRfaqozonecsv

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 59: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation PlotsPreliminariesPerforming the Simulation and Visualizing the ResultsExercise V

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 60: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Preliminaries

SimulationsPreliminaries The function outer()

1 x=56 y=13

2 outer(xy)

[1] [2] [3]

[1] 5 10 15

[2] 6 12 18

1 fcn lt-function(xy)z=x+y

2 outer(xyfcn)

[1] [2] [3]

[1] 6 7 8

[2] 7 8 9

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 61: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with persp()

Suppose we want to know what the function y times sin(x) looks like

1 Sample from the random uniform

2 x lt-sort(runif (100 min=0 max =10))

3 y lt- x+runif(1 min=0 max =20)

4 flt-function(x y)rlt-ysin(x)

5 zlt-outer(xyf)

6 persp(x y z col = lightblue shade = 01

ticktype = detailed expand =07)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 62: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with persp()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 63: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with image()

1 Format the data appropriate for the image ()

fcn

2 xseq lt-seq(from=range(x)[1] to=range(x)[2]

length =100)

3 yseq lt-seq(from=range(y)[1] to=range(y)[2]

length =100)

4 n1lt-length(xseq)

5 n2lt-length(yseq)

6 mat1 lt-matrix(z nrow=n1 ncol=n2 byrow=FALSE)

7 image(xseq yseq mat1)

8 to overlay a contour

9 contour(xseq yseq mat1 add=TRUE)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 64: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with image()

2 4 6 8

46

810

12

xseq

yse

q minus12

minus10

minus8

minus6

minus4

minus4

minus2

minus2

minus2

0

0

0

2

2

2

2 4

4

6 6

8 8

10 10

12 12

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 65: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IVisualize with plot3d()

Suppose we want to know what the same function y times sin(x) lookslike with a dynamic plot

1 Sample from the random uniform

2 x1 lt-runif (10000 min=0 max =10)

3 y1 lt- x1+runif (1000 min=0 max =20)

4 z1 lt- y1sin(x1)

5 plot3d(x1 y1 z1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 66: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Performing the Simulation and Visualizing the Results

Simulations IIVisualize with plot3d()

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 67: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Exercise V

Exercise V

Visualize the following function

z =sin(32x)

y(1)

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 68: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

1 Summary Plots

2 Time Series Plots

3 Geographical Plots

4 3D Plots

5 Simulation Plots

6 Useful Links for R

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 69: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Online Resources for R

Download R httpcranstatuclaedu

Search Engine for R http rseekorg

R Reference Cardhttpcranr-projectorgdoccontribShort-refcardpdf

R Graphics Galleryhttp researchstowers-instituteorgefgR

R Graph Gallery httpaddictedtorfreefrgraphiques

UCLA Statistics Information Portal http infostatuclaedugrad

UCLA Statistical Consulting Center http sccstatuclaedu

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 70: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Upcoming Mini-CourseThis Wednesday

500PM R Graphics III Customizing Graphics

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 71: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

We Want to Hear From You

Please complete our online survey Your feedback will help usimprove our mini-course series We greatly appreciate your time

http sccstatuclaedusurvey

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R
Page 72: R Graphics II: Graphics for Exploratory Data Analysisscc.stat.ucla.edu/page_attachments/0000/0185/Graphics_course.pdfSummary Plots Time Series Plots Geographical Plots 3D Plots Simulation

Summary Plots Time Series Plots Geographical Plots 3D Plots Simulation Plots Links

Thank youAny questions

Irina Kukuyeva ikukuyevastatuclaedu

R Graphics II Graphics for Exploratory Data Analysis UCLA SCC

  • Summary Plots
    • Basic Plots
    • Looking at Distributions
    • Identify-ing Observations
    • Exercise I
      • Time Series Plots
        • Multivariate Plots
        • Exercise II
          • Geographical Plots
            • Maps
            • Projection Maps
            • Exercise III
              • 3D Plots
                • lattice library
                • ggplot2 library
                • rgl library
                • Saving Plots as a PDF
                • Exercise IV
                  • Simulation Plots
                    • Preliminaries
                    • Performing the Simulation and Visualizing the Results
                    • Exercise V
                      • Useful Links for R