11 - 1 © 2003 pearson prentice hall chapter 11 simple linear regression

128
11 - 11 - 1 1 © 2003 Pearson Prentice © 2003 Pearson Prentice Hall Hall Chapter 11 Chapter 11 Simple Linear Regression Simple Linear Regression

Upload: johanna-stratton

Post on 01-Apr-2015

234 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 11

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Chapter 11Chapter 11

Simple Linear Regression Simple Linear Regression

Page 2: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 22

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Learning ObjectivesLearning Objectives

1.1. Describe the Linear Regression ModelDescribe the Linear Regression Model

2.2. State the Regression Modeling StepsState the Regression Modeling Steps

3.3. Explain Ordinary Least SquaresExplain Ordinary Least Squares

1.1. Understand and check model assumptionsUnderstand and check model assumptions

4.4. Compute Regression CoefficientsCompute Regression Coefficients

5.5. Predict Response VariablePredict Response Variable

6.6. Interpret Computer OutputInterpret Computer Output

Page 3: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 33

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ModelsModels

Page 4: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 44

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ModelsModels

1.1. Representation of Some PhenomenonRepresentation of Some Phenomenon

2.2. Mathematical Model Is a Mathematical Mathematical Model Is a Mathematical Expression of Some PhenomenonExpression of Some Phenomenon

3.3. Often Describe Relationships between Often Describe Relationships between VariablesVariables

4.4. TypesTypes Deterministic ModelsDeterministic Models Probabilistic ModelsProbabilistic Models

Page 5: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 55

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Deterministic Deterministic ModelsModels

1.1. Hypothesize Exact RelationshipsHypothesize Exact Relationships

2.2. Suitable When Prediction Error is Suitable When Prediction Error is NegligibleNegligible

3.3. Example: Force Is Exactly Example: Force Is Exactly Mass Times AccelerationMass Times Acceleration FF = = mm··aa

© 1984-1994 T/Maker Co.

Page 6: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 66

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Probabilistic ModelsProbabilistic Models

1.1. Hypothesize 2 ComponentsHypothesize 2 Components DeterministicDeterministic Random ErrorRandom Error

2.2. Example: Sales Volume Is 10 Times Example: Sales Volume Is 10 Times Advertising Spending + Random ErrorAdvertising Spending + Random Error YY = 10 = 10X X + + Random Error May Be Due to Factors Random Error May Be Due to Factors

Other Than AdvertisingOther Than Advertising

Page 7: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 77

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Probabilistic ModelsProbabilistic Models

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

Page 8: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 88

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression ModelsRegression Models

Page 9: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 99

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Probabilistic ModelsProbabilistic Models

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

Page 10: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1010

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression ModelsRegression Models

1.1. Answer ‘What Is the Relationship Answer ‘What Is the Relationship Between the Variables?’Between the Variables?’

2.2. Equation UsedEquation Used 1 Numerical Dependent (Response) Variable1 Numerical Dependent (Response) Variable

What Is to Be PredictedWhat Is to Be Predicted 1 or More Numerical or Categorical 1 or More Numerical or Categorical

Independent (Explanatory) VariablesIndependent (Explanatory) Variables

3.3. Used Mainly for Prediction & EstimationUsed Mainly for Prediction & Estimation

Page 11: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1111

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Specify Probability Distribution of Random Error TermRandom Error Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & Estimation Use Model for Prediction & Estimation

Page 12: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1212

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Model SpecificationModel Specification

Page 13: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1313

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Random Specify Probability Distribution of Random Error TermError Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & Estimation Use Model for Prediction & Estimation

Page 14: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1414

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Specifying the Specifying the ModelModel

1.1. Define VariablesDefine Variables

2.2. Hypothesize Nature of RelationshipHypothesize Nature of Relationship Expected Effects (i.e., Coefficients’ Signs)Expected Effects (i.e., Coefficients’ Signs) Functional Form (Linear or Non-Linear)Functional Form (Linear or Non-Linear) InteractionsInteractions

Page 15: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1515

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Model Specification Model Specification Is Based on TheoryIs Based on Theory

1.1. Theory of Field (e.g., Sociology)Theory of Field (e.g., Sociology)

2.2. Mathematical TheoryMathematical Theory

3.3. Previous ResearchPrevious Research

4.4. ‘Common Sense’‘Common Sense’

Page 16: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1616

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Advertising

Sales

Advertising

Sales

Advertising

Sales

Advertising

Sales

Advertising

Sales

Advertising

Sales

Advertising

Sales

Advertising

Sales

Thinking Challenge: Thinking Challenge: Which Is More Which Is More

Logical?Logical?

Page 17: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1717

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

Page 18: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1818

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

Page 19: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 1919

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

Simple

1 Explanatory1 ExplanatoryVariableVariable

Page 20: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2020

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

2+ Explanatory2+ ExplanatoryVariablesVariables

Simple Multiple

1 Explanatory1 ExplanatoryVariableVariable

Page 21: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2121

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

Linear

2+ Explanatory2+ ExplanatoryVariablesVariables

Simple Multiple

1 Explanatory1 ExplanatoryVariableVariable

Page 22: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2222

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

LinearNon-

Linear

2+ Explanatory2+ ExplanatoryVariablesVariables

Simple Multiple

1 Explanatory1 ExplanatoryVariableVariable

Page 23: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2323

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

LinearNon-

Linear

2+ Explanatory2+ ExplanatoryVariablesVariables

Simple Multiple

Linear

1 Explanatory1 ExplanatoryVariableVariable

Page 24: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2424

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

LinearNon-

Linear

2+ Explanatory2+ ExplanatoryVariablesVariables

Simple Multiple

Linear

1 Explanatory1 ExplanatoryVariableVariable

Non-Linear

Page 25: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2525

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Linear Regression Linear Regression ModelModel

Page 26: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2626

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Regression ModelsRegression Models

RegressionModels

LinearNon-

Linear

2+ ExplanatoryVariables

Simple

Non-Linear

Multiple

Linear

1 ExplanatoryVariable

RegressionModels

LinearNon-

Linear

2+ ExplanatoryVariables

Simple

Non-Linear

Multiple

Linear

1 ExplanatoryVariable

Page 27: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2727

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

Y = mX + b

b = Y-intercept

X

Changein Y

Change in X

m = Slope

Linear EquationsLinear Equations

High School TeacherHigh School Teacher© 1984-1994 T/Maker Co.

Page 28: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

YY XXii ii ii 00 11

Linear Regression Linear Regression ModelModel

1.1. Relationship Between Variables Is a Relationship Between Variables Is a Linear FunctionLinear Function

Dependent Dependent (Response) (Response) VariableVariable(e.g., income)(e.g., income)

Independent Independent (Explanatory) (Explanatory) Variable Variable (e.g., education)(e.g., education)

Population Population SlopeSlope

Population Population Y-InterceptY-Intercept

Random Random ErrorError

Page 29: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 2929

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Population & Population & Sample Regression Sample Regression

ModelsModels

Page 30: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3030

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Population & Population & Sample Regression Sample Regression

ModelsModels

PopulationPopulation

$ $

$

$

$

Page 31: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3131

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Population & Population & Sample Regression Sample Regression

ModelsModels

Unknown Relationship

PopulationPopulation

Y Xi i i 0 1

$

$

$

$ $

Page 32: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3232

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Population & Population & Sample Regression Sample Regression

ModelsModels

Unknown Relationship

PopulationPopulation Random SampleRandom Sample

Y Xi i i 0 1

$ $$

$

$ $$

$$ $$

Page 33: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3333

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Population & Population & Sample Regression Sample Regression

ModelsModels

Unknown Relationship

PopulationPopulation Random SampleRandom Sample

Y Xi i i 0 1

Y Xi i i 0 1Y Xi i i 0 1

$ $$

$

$ $$

$$ $$

Page 34: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3434

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

X

Y

X

Population Linear Population Linear Regression ModelRegression Model

Y Xi i i 0 1Y Xi i i 0 1

iXYE 10 iXYE 10

ObservedObservedvaluevalue

Observed valueObserved value

ii = Random error= Random error

Page 35: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3535

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

X

Y

X

Y Xi i i 0 1Y Xi i i 0 1

Sample Linear Sample Linear Regression ModelRegression Model

Y Xi i 0 1 Y Xi i 0 1

Unsampled Unsampled observationobservation

ii = Random = Random

errorerror

Observed valueObserved value

^

Page 36: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3636

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Estimating Parameters:Estimating Parameters:Least Squares MethodLeast Squares Method

Page 37: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3737

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Specify Probability Distribution of Random Error TermRandom Error Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & EstimationUse Model for Prediction & Estimation

Page 38: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3838

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

ScattergramScattergram

1.1. Plot of All (Plot of All (XXii, , YYii) Pairs) Pairs

2.2. Suggests How Well Model Will FitSuggests How Well Model Will Fit

Page 39: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 3939

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 40: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4040

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 41: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4141

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 42: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4242

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 43: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4343

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 44: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4444

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 45: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4545

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0204060

0 20 40 60

X

Y

Thinking ChallengeThinking Challenge

How would you draw a line through the How would you draw a line through the points? How do you determine which line points? How do you determine which line ‘fits best’?‘fits best’?

Page 46: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4646

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Least SquaresLeast Squares

1.1. ‘Best Fit’ Means Difference Between ‘Best Fit’ Means Difference Between Actual Y Values & Predicted Y Values Actual Y Values & Predicted Y Values Are a MinimumAre a Minimum ButBut Positive Differences Off-Set Negative Positive Differences Off-Set Negative

Page 47: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4747

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Least SquaresLeast Squares

1.1. ‘Best Fit’ Means Difference Between ‘Best Fit’ Means Difference Between Actual Y Values & Predicted Y Values Actual Y Values & Predicted Y Values Are a MinimumAre a Minimum ButBut Positive Differences Off-Set Negative Positive Differences Off-Set Negative

n

ii

n

iii YY

1

2

1

2ˆˆ

n

ii

n

iii YY

1

2

1

2ˆˆ

Page 48: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4848

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Least SquaresLeast Squares

1.1. ‘Best Fit’ Means Difference Between ‘Best Fit’ Means Difference Between Actual Y Values & Predicted Y Values Are Actual Y Values & Predicted Y Values Are a Minimuma Minimum ButBut Positive Differences Off-Set Negative Positive Differences Off-Set Negative

2.2. LS Minimizes the Sum of the Squared LS Minimizes the Sum of the Squared Differences (SSE)Differences (SSE)

n

ii

n

iii YY

1

2

1

2ˆˆ

n

ii

n

iii YY

1

2

1

2ˆˆ

Page 49: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 4949

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Least Squares Least Squares GraphicallyGraphically

2

Y

X

1 3

4

^^

^2

Y

X

1 3

4

^^

^^

Y X2 0 1 2 2 Y X2 0 1 2 2

Y Xi i 0 1 Y Xi i 0 1

LS minimizes ii

n2

112

22

32

42

LS minimizes ii

n2

112

22

32

42

Page 50: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5050

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient EquationsEquations

Sample SlopeSample Slope

Sample Y-interceptSample Y-intercept

Prediction EquationPrediction Equation

xy 10 ˆˆ

21

xx

yyxxSS

SS

i

ii

xx

xy

ii xy 10 ˆˆˆ

Page 51: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5151

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Computation TableComputation Table

Xi Yi Xi2 Yi

2 XiYi

X1 Y1 X12 Y1

2 X1Y1

X2 Y2 X22 Y2

2 X2Y2

: : : : :

Xn Yn Xn2 Yn

2 XnYn

XiYi

Xi2 Yi

2 XiYi

Xi Yi Xi2 Yi

2 XiYi

X1 Y1 X12 Y1

2 X1Y1

X2 Y2 X22 Y2

2 X2Y2

: : : : :

Xn Yn Xn2 Yn

2 XnYn

XiYi

Xi2 Yi

2 XiYi

Page 52: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5252

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Interpretation of Interpretation of CoefficientsCoefficients

Page 53: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5353

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Interpretation of Interpretation of CoefficientsCoefficients

1.1. Slope (Slope (11)) Estimated Estimated YY Changes by Changes by 11 for Each 1 for Each 1

Unit Increase in Unit Increase in XX If If 11 = 2, then Sales ( = 2, then Sales (YY) Is Expected to ) Is Expected to

Increase by 2 for Each 1 Unit Increase in Increase by 2 for Each 1 Unit Increase in Advertising (Advertising (XX))

^

^

^

Page 54: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5454

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Interpretation of Interpretation of CoefficientsCoefficients

1.1. Slope (Slope (11)) Estimated Estimated YY Changes by Changes by 11 for Each 1 for Each 1

Unit Increase in Unit Increase in XX If If 11 = 2, then Sales ( = 2, then Sales (YY) Is Expected to Increase ) Is Expected to Increase

by 2 for Each 1 Unit Increase in Advertising (by 2 for Each 1 Unit Increase in Advertising (XX))

2.2. Y-Intercept (Y-Intercept (00)) Average Value of Average Value of YY When When XX = 0 = 0

If If 00 = 4, then Average Sales ( = 4, then Average Sales (YY) Is Expected to ) Is Expected to Be 4 When Advertising (Be 4 When Advertising (XX) Is 0) Is 0

^

^

^^

^

Page 55: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5555

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation ExampleEstimation Example

You’re a marketing analyst for Hasbro Toys. You’re a marketing analyst for Hasbro Toys. You gather the following data:You gather the following data:

Ad $Ad $ Sales (Units)Sales (Units)11 1122 1133 2244 2255 44

What is the What is the relationshiprelationship between sales & advertising?between sales & advertising?

Page 56: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5656

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0

1

2

3

4

0 1 2 3 4 5

Scattergram Scattergram Sales vs. AdvertisingSales vs. Advertising

Sales

Advertising

Page 57: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5757

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Guess The Parameters!Guess The Parameters!

Page 58: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5858

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

0

1

2

3

4

0 1 2 3 4 5

Scattergram Scattergram Sales vs. AdvertisingSales vs. Advertising

Sales

Advertising

Page 59: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 5959

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation Solution Estimation Solution

TableTableXi Yi Xi

2 Yi2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Xi Yi Xi2 Yi

2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Page 60: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6060

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation SolutionEstimation Solution

10.0370.02ˆˆ

70.0

515

55

51015

37ˆ

10

2

1

2

12

11

11

XY

n

X

X

n

YX

YX

n

i

n

ii

i

n

ii

n

iin

iii

10.0370.02ˆˆ

70.0

515

55

51015

37ˆ

10

2

1

2

12

11

11

XY

n

X

X

n

YX

YX

n

i

n

ii

i

n

ii

n

iin

iii

Page 61: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6161

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

SolutionSolution

Page 62: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6262

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

SolutionSolution1.1. Slope (Slope (11))

Sales Volume (Sales Volume (YY) Is Expected to Increase ) Is Expected to Increase by .7 Units for Each $1 Increase in by .7 Units for Each $1 Increase in Advertising (Advertising (XX))

^

Page 63: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6363

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

SolutionSolution1.1. Slope (Slope (11))

Sales Volume (Sales Volume (YY) Is Expected to Increase ) Is Expected to Increase by .7 Units for Each $1 Increase in Advertising by .7 Units for Each $1 Increase in Advertising ((XX))

2.2. Y-Intercept (Y-Intercept (00)) Average Value of Sales Volume (Average Value of Sales Volume (YY) Is ) Is

-.10 Units When Advertising (-.10 Units When Advertising (XX) Is 0) Is 0 Difficult to Explain to Marketing ManagerDifficult to Explain to Marketing Manager Expect Some Sales Without AdvertisingExpect Some Sales Without Advertising

^

^

Page 64: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6464

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter EstimatesParameter Estimates

ParameterParameter Standard T for H0: Standard T for H0:

VariableVariable DF DF EstimateEstimate Error Param=0 Prob>|T| Error Param=0 Prob>|T|

INTERCEPINTERCEP 1 1 -0.1000-0.1000 0.6350 -0.157 0.8849 0.6350 -0.157 0.8849

ADVERTADVERT 1 1 0.70000.7000 0.1914 3.656 0.0354 0.1914 3.656 0.0354

Parameter Parameter Estimation Computer Estimation Computer

OutputOutput

0^ 1

^

k^

Page 65: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6565

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Derivation of Derivation of Parameter Parameter EquationsEquations

Goal: Minimize squared errorGoal: Minimize squared error

xnnyn

xy

xy

ii

iii

10

10

0

210

0

2

ˆˆ2

ˆˆ2

ˆˆˆ

ˆˆ

0

xy 10 ˆˆ

Page 66: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

Derivation of Derivation of Parameter Parameter EquationsEquations

iii

iii

iii

xxyyx

xyx

xy

11

10

1

210

1

2

ˆˆ2

ˆˆ2

ˆˆˆ

ˆˆ

0

xx

xy

iiii

iiii

SS

SS

yyxxxxxx

yyxxxx

1

1

1

ˆ

ˆ

ˆ

Page 67: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6767

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation Thinking Estimation Thinking

ChallengeChallengeYou’re an economist for the county You’re an economist for the county cooperative. You gather the following data:cooperative. You gather the following data:

Fertilizer (lb.)Fertilizer (lb.) Yield (lb.)Yield (lb.) 4 4 3.03.0 6 6 5.55.51010 6.56.51212 9.09.0

What is the What is the relationshiprelationship between fertilizer & crop yield?between fertilizer & crop yield?

© 1984-1994 T/Maker Co.

Page 68: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6868

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

02468

10

0 5 10 15

02468

10

0 5 10 15

Scattergram Scattergram Crop Yield vs. Crop Yield vs.

Fertilizer*Fertilizer*

Yield (lb.)Yield (lb.)Yield (lb.)Yield (lb.)

Fertilizer (lb.)Fertilizer (lb.)Fertilizer (lb.)Fertilizer (lb.)

Page 69: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 6969

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation Solution Estimation Solution

Table*Table*

Xi Yi Xi2 Yi

2 XiYi

4 3.0 16 9.00 12

6 5.5 36 30.25 33

10 6.5 100 42.25 65

12 9.0 144 81.00 108

32 24.0 296 162.50 218

Xi Yi Xi2 Yi

2 XiYi

4 3.0 16 9.00 12

6 5.5 36 30.25 33

10 6.5 100 42.25 65

12 9.0 144 81.00 108

32 24.0 296 162.50 218

Page 70: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7070

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Parameter Parameter Estimation Solution*Estimation Solution*

80.0865.06ˆˆ

65.0

432

296

42432

218ˆ

10

2

1

2

12

11

11

XY

n

X

X

n

YX

YX

n

i

n

ii

i

n

ii

n

iin

iii

80.0865.06ˆˆ

65.0

432

296

42432

218ˆ

10

2

1

2

12

11

11

XY

n

X

X

n

YX

YX

n

i

n

ii

i

n

ii

n

iin

iii

Page 71: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7171

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

Solution*Solution*

Page 72: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7272

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

Solution*Solution*

1.1. Slope (Slope (11)) Crop Yield (Crop Yield (YY) Is Expected to Increase ) Is Expected to Increase

by .65 lb. for Each 1 lb. Increase in Fertilizer by .65 lb. for Each 1 lb. Increase in Fertilizer ((XX))

^

Page 73: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7373

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient Coefficient Interpretation Interpretation

Solution*Solution*

1.1. Slope (Slope (11)) Crop Yield (Crop Yield (YY) Is Expected to Increase ) Is Expected to Increase

by .65 lb. for Each 1 lb. Increase in Fertilizer by .65 lb. for Each 1 lb. Increase in Fertilizer ((XX))

2.2. Y-Intercept (Y-Intercept (00)) Average Crop Yield (Average Crop Yield (YY) Is Expected to Be ) Is Expected to Be

0.8 lb. When No Fertilizer (0.8 lb. When No Fertilizer (XX) Is Used) Is Used

^

^

Page 74: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7474

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Probability Distribution Probability Distribution

of Random Errorof Random Error

Page 75: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7575

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Specify Probability Distribution of Random Error TermRandom Error Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & Estimation Use Model for Prediction & Estimation

Page 76: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7676

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Linear Regression Linear Regression Assumptions Assumptions

1.1. Mean of Probability Distribution of Mean of Probability Distribution of Error Is 0Error Is 0

2.2. Probability Distribution of Error Has Probability Distribution of Error Has Constant VarianceConstant Variance1.1. Exercise: Constant across what?Exercise: Constant across what?

3.3. Probability Distribution of Error is Probability Distribution of Error is NormalNormal

4.4. Errors Are Independent Errors Are Independent

Page 77: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7777

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Error Error Probability Probability DistributionDistribution

Y

f()

X

X 1X 2

Y

f()

X

X 1X 2

^

Page 78: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7878

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Random Error Random Error VariationVariation

Page 79: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 7979

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Random Error Random Error VariationVariation

1.1. Variation of Actual Variation of Actual YY from Predicted from Predicted YY

Page 80: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8080

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Random Error Random Error VariationVariation

1.1. Variation of Actual Variation of Actual YY from Predicted from Predicted YY

2.2. Measured by Standard Error of Measured by Standard Error of Regression ModelRegression Model Sample Standard Deviation of Sample Standard Deviation of , , ss^

Page 81: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8181

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Random Error Random Error VariationVariation

1.1. Variation of Actual Variation of Actual YY from Predicted from Predicted YY

2.2. Measured by Standard Error of Measured by Standard Error of Regression ModelRegression Model Sample Standard Deviation of Sample Standard Deviation of , , ss

3. 3. Affects Several FactorsAffects Several Factors Parameter SignificanceParameter Significance Prediction AccuracyPrediction Accuracy

^

Page 82: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8282

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Evaluating the ModelEvaluating the Model

Testing for SignificanceTesting for Significance

Page 83: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8383

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Specify Probability Distribution of Random Error TermRandom Error Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & EstimationUse Model for Prediction & Estimation

Page 84: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8484

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test of Slope Test of Slope CoefficientCoefficient

1.1. Shows If There Is a Linear Relationship Shows If There Is a Linear Relationship Between Between XX & & YY

2.2. Involves Population Slope Involves Population Slope 11

3.3. Hypotheses Hypotheses HH00: : 1 1 = 0 (No Linear Relationship) = 0 (No Linear Relationship)

HHaa: : 11 0 (Linear Relationship) 0 (Linear Relationship)

4.4. Theoretical Basis Is Sampling Distribution Theoretical Basis Is Sampling Distribution of Slopeof Slope

Page 85: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8585

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Sampling Sampling Distribution Distribution

of Sample Slopesof Sample Slopes

Page 86: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8686

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

Population LineX

Sample 1 Line

Sample 2 Line

Y

Population LineX

Sample 1 Line

Sample 2 Line

Sampling Sampling Distribution Distribution

of Sample Slopesof Sample Slopes

Page 87: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8787

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

Population LineX

Sample 1 Line

Sample 2 Line

Y

Population LineX

Sample 1 Line

Sample 2 Line

Sampling Sampling Distribution Distribution

of Sample Slopesof Sample Slopes

All Possible All Possible Sample SlopesSample Slopes

Sample 1:Sample 1: 2.52.5

Sample 2:Sample 2: 1.6 1.6

Sample 3:Sample 3: 1.81.8

Sample 4:Sample 4: 2.12.1 : : : :Very large number of Very large number of sample slopessample slopes

Page 88: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8888

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

Population LineX

Sample 1 Line

Sample 2 Line

Y

Population LineX

Sample 1 Line

Sample 2 Line

Sampling Sampling Distribution Distribution

of Sample Slopesof Sample Slopes

11

All Possible All Possible Sample SlopesSample Slopes

Sample 1:Sample 1: 2.52.5

Sample 2:Sample 2: 1.6 1.6

Sample 3:Sample 3: 1.81.8

Sample 4:Sample 4: 2.12.1 : : : :Very large number of Very large number of sample slopessample slopes

Sampling DistributionSampling Distribution

11

11SS

^

^

Page 89: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 8989

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Slope Coefficient Slope Coefficient Test StatisticTest Statistic

n

X

X

SS

St

n

iin

ii

n

2

1

1

2

ˆ

ˆ

112

1

1

where

ˆ

n

X

X

SS

St

n

iin

ii

n

2

1

1

2

ˆ

ˆ

112

1

1

where

ˆ

Page 90: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9090

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test of Slope Test of Slope Coefficient ExampleCoefficient Example

You’re a marketing analyst for Hasbro Toys. You’re a marketing analyst for Hasbro Toys. You find You find bb00 = -.1 = -.1,, bb11 = .7 = .7 & & ss = .60553= .60553..

Ad $Ad $ Sales (Units)Sales (Units)11 1122 1133 2244 2255 44

Is the relationship Is the relationship significantsignificant at the at the .05.05 level? level?

Page 91: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9191

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Solution TableSolution Table

Xi Yi Xi2 Yi

2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Xi Yi Xi2 Yi

2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Page 92: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9292

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test of Slope Test of Slope Parameter Parameter

SolutionSolutionHH00: : 11 = 0 = 0

HHaa: : 11 0 0

.05.05

df df 5 - 2 = 35 - 2 = 3

Critical Value(s):Critical Value(s):

Test Statistic: Test Statistic:

Decision:Decision:

Conclusion:Conclusion:

t0 3.1824-3.1824

.025

Reject Reject

.025

t0 3.1824-3.1824

.025

Reject Reject

.025

tS

.

..

1 1

1

0 70 001915

3 656tS

.

..

1 1

1

0 70 001915

3 656

Reject at Reject at = .05 = .05

There is evidence of a There is evidence of a relationshiprelationship

Page 93: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9393

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test StatisticTest StatisticSolutionSolution

1915.0

515

55

60553.0

where

656.31915.0

070.0ˆ

32

1

1

2

ˆ

ˆ

112

1

1

n

X

X

SS

St

n

iin

ii

n

1915.0

515

55

60553.0

where

656.31915.0

070.0ˆ

32

1

1

2

ˆ

ˆ

112

1

1

n

X

X

SS

St

n

iin

ii

n

Page 94: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9494

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test of Slope Test of Slope ParameterParameter

Computer OutputComputer Output Parameter EstimatesParameter Estimates

Parameter Standard Parameter Standard T for H0:T for H0:

VariableVariable DF Estimate Error DF Estimate Error Param=0 Prob>|T|Param=0 Prob>|T|

INTERCEP 1 -0.1000 0.6350 -0.157 0.8849INTERCEP 1 -0.1000 0.6350 -0.157 0.8849

ADVERTADVERT 1 0.7000 0.1914 1 0.7000 0.1914 3.6563.656 0.03540.0354

t = k / S

P-Value

Skk k

^^^^

Page 95: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9595

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Measures of Measures of Variation Variation

in Regression in Regression 1.1. Total Sum of Squares (SSTotal Sum of Squares (SSyyyy))

Measures Variation of Observed Measures Variation of Observed YYii Around the MeanAround the MeanYY

2.2. Explained Variation (SSR)Explained Variation (SSR) Variation Due to Relationship Between Variation Due to Relationship Between

XX & & YY

3.3. Unexplained VariationUnexplained Variation (SSE) (SSE) Variation Due to Other FactorsVariation Due to Other Factors

Page 96: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9696

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

X

Y

X i

Y

X

Y

X i

Variation MeasuresVariation Measures

Y Xi i 0 1 Y Xi i 0 1

Total sum Total sum

of squares of squares

(Y(Yii - -Y)Y)22

Unexplained sum Unexplained sum

of squares (Yof squares (Yii - -

YYii))22

^

Explained sum of Explained sum of

squares (Ysquares (Yii - -Y)Y)22 ^

YYii

Page 97: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9797

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

1.1. ProportionProportion of Variation ‘Explained’ by of Variation ‘Explained’ by Relationship Between Relationship Between XX & & YY

Coefficient of Coefficient of DeterminationDetermination

n

ii

n

ii

n

ii

YY

YYYY

r

1

2

1

2

1

2

2

ˆ

Variation Total

Variation Explained

n

ii

n

ii

n

ii

YY

YYYY

r

1

2

1

2

1

2

2

ˆ

Variation Total

Variation Explained

0 r2 1

Page 98: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9898

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Y

X

Y

X

Y

X

Coefficient of Coefficient of Determination Determination

ExamplesExamplesY

X

r2 = 1 r2 = 1

r2 = .8 r2 = 0

Page 99: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 9999

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Coefficient of Determination Determination

ExampleExampleYou’re a marketing analyst for Hasbro You’re a marketing analyst for Hasbro

Toys. You find Toys. You find 00 = -0.1 & = -0.1 & 11 = 0.7. = 0.7.

Ad $Ad $ Sales (Units)Sales (Units)11 1122 1133 2244 2255 44

Interpret a Interpret a coefficient of coefficient of determination determination ofof 0.8167.0.8167.

^^

Page 100: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 100100

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

r r 22 Computer Output Computer Output

Root MSE 0.60553Root MSE 0.60553 R-square 0.8167R-square 0.8167

Dep Mean 2.00000 Dep Mean 2.00000 Adj R-sq 0.7556Adj R-sq 0.7556

C.V. 30.27650 C.V. 30.27650

r2 adjusted for number of explanatory variables & sample size

S

r2

Page 101: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 101101

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Using the Model for Using the Model for Prediction & EstimationPrediction & Estimation

Page 102: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 102102

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Regression Modeling Regression Modeling Steps Steps

1.1. Hypothesize Deterministic ComponentHypothesize Deterministic Component

2.2. Estimate Unknown Model ParametersEstimate Unknown Model Parameters

3.3. Specify Probability Distribution of Specify Probability Distribution of Random Error TermRandom Error Term Estimate Standard Deviation of ErrorEstimate Standard Deviation of Error

4.4. Evaluate ModelEvaluate Model

5.5. Use Model for Prediction & Estimation Use Model for Prediction & Estimation

Page 103: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 103103

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Prediction With Prediction With Regression ModelsRegression Models

1.1. Types of PredictionsTypes of Predictions Point EstimatesPoint Estimates Interval EstimatesInterval Estimates

2.2. What Is PredictedWhat Is Predicted Population Mean Response Population Mean Response EE((YY) for ) for

Given Given XX Point on Population Regression LinePoint on Population Regression Line

Individual Response (Individual Response (YYii) for Given ) for Given XX

Page 104: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 104104

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

What Is PredictedWhat Is Predicted

Mean Y, E(Y)

Y

Y i= 0

+ 1X

^Y Individual

Prediction, Y

E(Y) = 0 + 1X

^

XXP

^^

Mean Y, E(Y)

Y

Y i= 0

+ 1X

^Y Individual

Prediction, Y

E(Y) = 0 + 1X

^

XXP

^^

Page 105: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 105105

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ConfidenceConfidence Interval Interval Estimate of Mean Estimate of Mean YY

n

ii

p

Y

YnYn

XX

XX

nSS

StYYEStY

1

2

2

ˆ

ˆ2/,2ˆ2/,2

1

where

ˆ)(ˆ

n

ii

p

Y

YnYn

XX

XX

nSS

StYYEStY

1

2

2

ˆ

ˆ2/,2ˆ2/,2

1

where

ˆ)(ˆ

Page 106: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 106106

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Factors Affecting Factors Affecting Interval WidthInterval Width

1.1. Level of Confidence (1 - Level of Confidence (1 - )) Width Increases as Confidence IncreasesWidth Increases as Confidence Increases

2.2. Data Dispersion (Data Dispersion (ss)) Width Increases as Variation IncreasesWidth Increases as Variation Increases

3.3. Sample SizeSample Size Width Decreases as Sample Size IncreasesWidth Decreases as Sample Size Increases

4.4. Distance of Distance of XXpp from Mean from MeanXX Width Increases as Distance IncreasesWidth Increases as Distance Increases

Page 107: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 107107

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Why Distance from Why Distance from Mean?Mean?

Sample 2 Line

Y

XX1 X2

Y_ Sample 1 Line

Sample 2 Line

Y

XX1 X2

Y_ Sample 1 Line

Greater Greater dispersion dispersion than than XX11

XX

Page 108: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 108108

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ConfidenceConfidence Interval Interval Estimate ExampleEstimate Example

You’re a marketing analyst for Hasbro Toys. You’re a marketing analyst for Hasbro Toys. You find You find bb00 = -.1 = -.1,, bb11 = .7 = .7 & & ss = .60553= .60553..

Ad $Ad $ Sales (Units)Sales (Units)11 1122 1133 2244 2255 44

Estimate the Estimate the meanmean sales when sales when advertising is advertising is $4$4 at the at the .05.05 level. level.

Page 109: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 109109

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Solution TableSolution Table

Xi Yi Xi2 Yi

2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Xi Yi Xi2 Yi

2 XiYi

1 1 1 1 1

2 1 4 1 2

3 2 9 4 6

4 2 16 4 8

5 4 25 16 20

15 10 55 26 37

Page 110: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 110110

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ConfidenceConfidence Interval Interval Estimate SolutionEstimate Solution

7553.3)(6445.1

3316.01824.37.2)(3316.01824.37.2

3316.010

34

5

160553.

7.247.01.0ˆ

ˆ)(ˆ

2

ˆ

ˆ2/,2ˆ2/,2

YE

YE

S

Y

StYYEStY

Y

YnYn

7553.3)(6445.1

3316.01824.37.2)(3316.01824.37.2

3316.010

34

5

160553.

7.247.01.0ˆ

ˆ)(ˆ

2

ˆ

ˆ2/,2ˆ2/,2

YE

YE

S

Y

StYYEStY

Y

YnYn

XX to be predicted to be predictedXX to be predicted to be predicted

Page 111: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 111111

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

n

ii

PYY

YYnPYYn

XX

XX

nSS

StYYStY

1

2

2

ˆ

ˆ2/,2ˆ2/,2

11

where

ˆˆ

n

ii

PYY

YYnPYYn

XX

XX

nSS

StYYStY

1

2

2

ˆ

ˆ2/,2ˆ2/,2

11

where

ˆˆ

PredictionPrediction Interval Interval of Individual of Individual

ResponseResponse

Note!Note!

Page 112: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 112112

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Why the Extra ‘SWhy the Extra ‘S’’??

Expected(Mean) Y

Y

Y i= 0

+ 1X i

^

Y we're trying to predict

Prediction, Y

E(Y) = 0 + 1X

^

XXP

^

^Expected(Mean) Y

Y

Y i= 0

+ 1X i

^

Y we're trying to predict

Prediction, Y

E(Y) = 0 + 1X

^

XXP

^

^

Page 113: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 113113

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Interval Estimate Interval Estimate Computer OutputComputer Output

Dep Var Pred Std Err Dep Var Pred Std Err Low95% Upp95% Low95% Upp95%Low95% Upp95% Low95% Upp95%

Obs SALES Value Predict Obs SALES Value Predict Mean Mean Predict PredictMean Mean Predict Predict

1 1.000 0.600 0.469 -0.892 2.092 -1.837 3.037 1 1.000 0.600 0.469 -0.892 2.092 -1.837 3.037

2 1.000 1.300 0.332 0.244 2.355 -0.897 3.4972 1.000 1.300 0.332 0.244 2.355 -0.897 3.497

3 2.000 2.000 0.271 1.138 2.861 -0.111 4.1113 2.000 2.000 0.271 1.138 2.861 -0.111 4.111

4 2.000 4 2.000 2.700 0.332 1.644 3.755 0.502 4.897 2.700 0.332 1.644 3.755 0.502 4.897

5 4.000 3.400 0.469 1.907 4.892 0.962 5.8375 4.000 3.400 0.469 1.907 4.892 0.962 5.837

Predicted Predicted YY when when XX = 4 = 4

Confidence Confidence IntervalInterval

SSYYPrediction Prediction IntervalInterval

Page 114: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 114114

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Hyperbolic Interval Hyperbolic Interval BandsBands

X

Y

X

Y i= 0

+ 1X i

^

XP

_

^^

X

Y

X

Y i= 0

+ 1X i

^

XP

_

^^

Page 115: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 115115

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Correlation ModelsCorrelation Models

Page 116: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 116116

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Types of Types of Probabilistic ModelsProbabilistic Models

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

ProbabilisticModels

RegressionModels

CorrelationModels

OtherModels

Page 117: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 117117

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Correlation ModelsCorrelation Models

1.1. Answer ‘Answer ‘How Strong How Strong Is the Linear Is the Linear Relationship Between 2 Variables?’Relationship Between 2 Variables?’

2.2. Coefficient of Correlation UsedCoefficient of Correlation Used Population Correlation Coefficient Denoted Population Correlation Coefficient Denoted

(Rho) (Rho) Values Range from -1 to +1Values Range from -1 to +1 Measures Degree of AssociationMeasures Degree of Association

3.3. Used Mainly for UnderstandingUsed Mainly for Understanding

Page 118: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 118118

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

1.1. Pearson Product Moment Coefficient of Pearson Product Moment Coefficient of Correlation, Correlation, rr::

Sample Coefficient Sample Coefficient of Correlationof Correlation

n

ii

n

ii

n

iii

YYXX

YYXX

r

1

2

1

2

1

ionDeterminat oft Coefficien

n

ii

n

ii

n

iii

YYXX

YYXX

r

1

2

1

2

1

ionDeterminat oft Coefficien

Page 119: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 119119

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000-.5-.5 +.5+.5

Page 120: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 120120

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000-.5-.5 +.5+.5

No No CorrelationCorrelation

Page 121: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 121121

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000

Increasing degree of Increasing degree of negative correlationnegative correlation

-.5-.5 +.5+.5

No No CorrelationCorrelation

Page 122: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 122122

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000-.5-.5 +.5+.5

Perfect Perfect Negative Negative

CorrelationCorrelationNo No

CorrelationCorrelation

Page 123: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 123123

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000-.5-.5 +.5+.5

Perfect Perfect Negative Negative

CorrelationCorrelationNo No

CorrelationCorrelation

Increasing degree of Increasing degree of positive correlationpositive correlation

Page 124: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 124124

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Correlation Coefficient of Correlation ValuesValues

-1.0-1.0 +1.0+1.000

Perfect Perfect Positive Positive

CorrelationCorrelation

-.5-.5 +.5+.5

Perfect Perfect Negative Negative

CorrelationCorrelationNo No

CorrelationCorrelation

Page 125: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 125125

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Coefficient of Coefficient of CorrelationCorrelation ExamplesExamples

Y

X

Y

X

Y

X

Y

X

r = 1 r = -1

r = .89 r = 0

Page 126: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 126126

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

Test of Test of Coefficient of Coefficient of Correlation Correlation

1.1. Shows If There Is a Linear Relationship Shows If There Is a Linear Relationship Between 2 Numerical VariablesBetween 2 Numerical Variables

2.2. Same Conclusion as Testing Same Conclusion as Testing Population Slope Population Slope 11

3.3. Hypotheses Hypotheses HH00: : = 0 (No Correlation) = 0 (No Correlation)

HHaa: : 0 (Correlation) 0 (Correlation)

Page 127: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

11 - 11 - 127127

© 2003 Pearson Prentice Hall© 2003 Pearson Prentice Hall

ConclusionConclusion

1.1. Described the Linear Regression ModelDescribed the Linear Regression Model

2.2. Stated the Regression Modeling StepsStated the Regression Modeling Steps

3.3. Explained Ordinary Least SquaresExplained Ordinary Least Squares

4.4. Computed Regression CoefficientsComputed Regression Coefficients

5.5. Predicted Response VariablePredicted Response Variable

6.6. Interpreted Computer OutputInterpreted Computer Output

Page 128: 11 - 1 © 2003 Pearson Prentice Hall Chapter 11 Simple Linear Regression

End of Chapter

Any blank slides that follow are blank intentionally.