ams572_notes_12

# ams572_notes_12 - AMS 572 Lecture Notes#12 Nov 4th 2010 Ch...

This preview shows pages 1–4. Sign up to view the full content.

AMS 572 Lecture Notes #12 Nov. 4 th , 2010 Ch. 9. Categorical Data Analysis Exact test on one population proportion: Data: sample of n, X are # of “Successes”, n-X are # of “Failures” ~ ( , ) X B n p ( ) (1 ) , 0,1,2, , x n x n P X x p p x n x -   = = - =     K (1)  0 0 0 : : a H p p H p p = 0 0 0 0 ( | : ) (1 ) n i n i i x n p value P X x H p p p p i - =   - = = = -     (2)  0 0 0 : : a H p p H p p = < 0 0 0 0 0 ( | : ) (1 ) x i n i i n p value P X x H p p p p i - =   - = = = -     (3)  0 0 0 : : a H p p H p p = 0 0 2*min{ ( | ), ( | )} p value P X x H P X x H - = . 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Inference on Two Population Proportions: 1 p , 2 p Independent samples, both are large e.g. Suppose we wish to compare the proportions of smokers among male and female students in SBU. Two large independent samples: Population 1: 1 p Population 2: 2 p Sample 1: 1 1 1 1 , , n x n x - Sample 2: 2 2 2 2 , , n x n x - ( 1 1 1 5, 5 x n x - , 2 2 2 5, 5 x n x - ) 1 Point estimator: · µ 1 2 1 2 1 2 1 2 x x p p p p n n - = - = - 2 By CLT, µ 1 1 1 1 1 (1 ) ( , ) p p p N p n - : 2 2 2 2 2 (1 ) ( , ) p p p N p n - : Q Two samples are independent µ 1 1 2 2 1 2 1 2 1 2 (1 ) (1 ) ( , ) p p p p p p N p p n n - - - - + : 3 P.Q. for 1 2 p p - : µ 1 2 1 2 1 1 2 2 1 2 ( ) (0,1) (1 ) (1 ) p p p p Z N p p p p n n - - - = - - + : not P.Q. µ µ µ * 1 2 1 2 1 1 2 2 1 2 ( ) (0,1) (1 ) (1 ) p p p p Z N p p p p n n - - - = - - + : P.Q. 4 100(1- α )% large samples CI for 1 2 p p - : 2
µ * 1 2 1 2 1 2 2 2 2 2 1 ( ) ( ) P Z Z Z P p p Z S p p p p Z S α - = - = - - - - + Here, µ µ 1 1 2 2 1 2 (1 ) (1 ) p p p p S n n - - = + 5 Test General case: 0 1 2 1 2 : : a H p p H p p - = ∆ - µ µ µ 0 1 2 0 0 1 1 2 2 1 2 : (0,1) (1 ) (1 ) H p p T Z N p p p p n n - - ∆ = - - + : p-value=P(Z 0 Z ) At the significance level , we reject 0 H if 0 Z Z and p-value< α . When =0, one often uses the following test statistic µ µ µ 0 * 1 2 1 2 0 (0,1) 1 1 (1 )( ) H p p Z N p p n n - - = - + : here µ µ 1 1 2 2 1 2 1 2 1 2 n p n p X X p n n n n + + = = + + Example. A random sample of Democrats and a random sample of Republicans  were polled on an issue. Of 200 Republicans, 90 would vote yes on the issue; of  100 democrats, 58 would vote yes. Let p 1  and p 2  denote respectively the  proportions of all Democrats or all Republicans who would vote yes on this issue.  (a) Construct a 95% confidence interval for (p 1  - p 2 ) (b) Can we say that more Democrats than Republicans favor the issue at the  1% level of significance? Please report the p-value.  (c) Please write up the entire SAS program necessary to answer question  raised in (b). Please include the data step. 3

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 01/31/2011 for the course AMS 572 taught by Professor Weizhu during the Fall '10 term at SUNY Stony Brook.

### Page1 / 17

ams572_notes_12 - AMS 572 Lecture Notes#12 Nov 4th 2010 Ch...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online