This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Midterm Solution : 1 a) Since we choose the different days on which TSE went up and down, we should use twosample test. By looking at the given boxplots ( difference boxplot should not be taken into consideration) , we can assume that the samples are from normally distributed populations. We can use twosample ttest. Second assumption would be the equal variances. If we find equal variances, we would use pooled standard deviation, if not we will go with the other formula. : H The two population variances are equal : ( 29 2 2 2 1 = : A H The two population variances are not equal : ( 29 2 2 2 1 70 . 1 94 . 60 . 1 971 . 267 . 1 2 2 2 1 2 2 = = = = s s F 98 . 1 05 . 2 2245 = F F , df = 23, 23 (since we dont have a value for 23, 23 we will take the nearest value which is for 24, 24.) < 2 F F Do not reject H and assume equal variances. b) : H There is no difference in the populations C.D rates when the TSE goes up vs. goes down : A H There is difference in the populations C.D rates down up H = : down up A H : c) 90 % C.I: 2 1 2 2 1 1 1 n n s t x x p +  ( 29 ( 29 = + + = 2 1 1 2 1 2 2 2 2 1 1 n n s n s n s p pooled standard deviation...
View
Full
Document
This note was uploaded on 09/27/2008 for the course ADM 2304 taught by Professor Unknown during the Summer '04 term at University of Ottawa.
 Summer '04
 Unknown

Click to edit the document details