This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 9.1 z Test and Confidence intervals for a Difference Between two population means Let μ 1 , σ 1 and μ 2 , σ 2 are the mean and standard deviation of two populations or processes. Let n, x and s 1 represent the sample size, sample mean and sample s.d. from the first population, and m, y and s 2 represent the sample size, sample mean and sample s.d. from the second population. The properties: Confidence intervals for μ 1 μ 2 (When two different processes or populations are compared) 1 1. If both population distributions are normal, so is the sampling distribution of y x 2. If both sizes n and m are large, then by Central Limit Theorem y x is approximately normal 3. The standard deviation is m n y x 2 2 2 1 σ σ σ + = 4. y x is unbiased estimator for estimating μ 1 μ 2 Do Exercise 2/page334: Assumption 1. Both samples are large and independent on each other 2. Sample sizes may differ ) (1 100 α % confidence interval for μ 1 μ 2 is = 2 1 c α m s n s z y x c 2 2 2 1 + ± 2 Example : Use the accompanying data to estimate with a 95% confidence interval the difference between true average compressive strength (N/mm 2 ) for 7 – dayold concrete specimens and true average strength for 28dayold specimens. 7day old: 1 n = 68 1 x = 26.99 1 s = 4.89 28day old: 2 n = 74 2 x = 35.76 2 s = 6.43 Test whether there is a difference between the means…. Solution : A 95% confidence interval for the difference between the true average compressive strength for 7dayold concrete specimens and the true average strength for 28dayold concrete specimens is: ( 29 ( 29 ( 29 ( 29 ( 29 ( 29 9 . 6 , 64 . 10 87 . 1 77 . 8 74 43 . 6 68 89 . 4 96 . 1 76 . 35 99 . 26 96 . 1 2 2 2 2 2 1 2 1 ± + ± + ± m s n s x x HW assignment: 3 Section 9.1: #3, 5ab, 7, 9, 11, 8* 9.2 TwoSample t Test and CI for 2 1 μ μ of Two Normal Populations We are interested in comparing two normal populations and in obtaining the CI for the difference in their true means. • Sampling designs: independent vs. paired • Confidence intervals for independent sample • Confidence Intervals for matched samples • Two independent samples n 1 observations n 2 observations x 1 : x 11, x 12, … x 2 : x 21, x 21, … ( The observations are independent). 4 μ 1 σ 1 μ 2 σ 2 Independent Samples : The sample values ) , ( 2 1 i i x x selected from one population are not related and somehow paired with the sample values selected from the other population. Dependent Samples : If the values in one sample are related to the values in the other sample, the samples are dependent . Such samples are often referred to as matched pairs or paired samples . Example: Independent samples: one group of subjects is treated with a drug, and a separate group is given a placebo....
View
Full
Document
This note was uploaded on 07/25/2008 for the course STT 430 taught by Professor Nane during the Spring '08 term at Michigan State University.
 Spring '08
 NANE

Click to edit the document details