hypergeometric distributions
DESCRIPTION
Hypergeometric Distributions. Remember, for rolling dice. uniform. binomial. geometric. Rolling a pair P(pair) = eventually. Rolling a 4 P(4) = 1/6. Rolling a 7 P(pair) = 6/36. A Hypergeometric distribution does not fit any of these models. It is close to the Binomial Model. - PowerPoint PPT PresentationTRANSCRIPT
Hypergeometric Distributions
Remember, for rolling dice
uniform
Rolling a 4
P(4) = 1/6
binomial
Rolling a 7
P(pair) = 6/36
geometricRolling a pairP(pair) = eventually
A Hypergeometric distribution does not fit any of these
models.
It is close to the Binomial Model.
HD are used when the probabilities of successive trials (or selections)
are dependent
Meeting the first condition determines the probability of the second.
The random variable (X) is the number of successful trials in an experiment.For efficiency, we will use combinations to help us count
Jury SelectionA six member jury is needed.
There are 10 women and 6 men available.
Create a probability distribution for all the possible juries.
Identify the DRV X as the number of women on the jury.
Try a binomial model….
Jury SelectionA six member jury is needed.
There are 10 women and 6 men available.
Create a probability distribution for all the possible juries.
X: a womanp = 10 / 16q = 6 /16
Suppose we select 4 women, so x = 4There is a breakdown when we model this…..not the same as dice….not all the probabilities stay the same
Jury SelectionA six member jury is needed.
There are 10 women and 6 men available.
Create a probability distribution for all the possible juries.
X: a womanp = 10 / 16q = 6 /16
Suppose we select 4 women, so x = 4Our way around this is to chose the women in groups (like subsets, we have done this before….)
Determine the PD for the number of women selected for a 6 member jury
from a group of10 women and 8 men
Let (the RV) X be the number of women selected.
Number of women (x) Probability P(x)
P(x) = n(x)
n(S)
= women X men
Total number of
selections
= 10Cx X 8C6 - x
18C6
Probability in a Hypergeometric Distribution
P(x) = aCx X n-aCr-x
nCrx: number of successful selections
r: size of the subset drawn (6)
a: total number of successful elements to draw from (10)
n: all selections (18)
Expected ValueE(X) = r a
nWhat is the expected number of women on the committee?
r = 6
a = 10
n = 18E(X) = 6[10/18]
= 3.3
Homework
Pg 404 1,3,7,11,12
Remember
1. If you roll a die 4 times, what is the probability that
a) you roll a 3 twice?
b) you roll a 3 on the last roll?