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: (a) P { T= 2 } (b) P { T= 5 } (c) P { T 3 } 4. In order to test the eciency of some new statistical software, it was tested against the old software using some real data. Each program was run on 10 data sets and timed to see how long it would take to produce the output. Test whether or not the new software runs signicantly faster using a signedrank test. Use = 0 . 05. Data set Old program New Program 1 71.6 79.9 2 35.5 15.2 3 37.7 25.2 4 35.4 21.0 5 37.5 27.3 6 75.8 67.1 7 41.6 36.3 8 80.1 98.5 9 61.3 87.9 10 37.9 32.2 5. For the 300 observations in the data set to test H : = 0 versus H A : 6 = 0 where is the median of the distribution, we calculated the signedrank statistics T= 25 , 850. 1 (a) Calculate T + . (b) Calculate E T when the null hypothesis is true. (c) Is there sucient evidence to reject the null hypothesis at a = 0 . 05? Calculate the appropriate P value. 2...
View
Full
Document
This note was uploaded on 10/02/2009 for the course PSTAT 5E taught by Professor Eduardomontoya during the Spring '08 term at UCSB.
 Spring '08
 EduardoMontoya

Click to edit the document details