|
Activity Number:
|
123
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304798 |
|
Title:
|
Consonant Closed Test Procedures in Dose-Response Study
|
|
Author(s):
|
Bushi Wang*+ and Xinping Cui
|
|
Companies:
|
University of California, Riverside and University of California, Riverside
|
|
Address:
|
3384 Idaho St, Riverside, CA, 92507,
|
|
Keywords:
|
multiple testing procedures ; coherence ; consonance ; closed test
|
|
Abstract:
|
Coherence and Consonance are two important concepts in multiple testing procedures. Closed test is a well-known coherent hypothesis testing procedure which is not consonant in general. In this paper, we propose two consonant closed tests for the clinical trial dose-response study which satisfy both coherence and consonance requirement. A simulation study shows that our tests are more powerful than Marcus et al. (1976)'s closed test and as powerful as Hommel (1988)'s test. Our new consonant closed tests work better in identifying false null hypotheses when the majority of them are true. They are powerful tools in dose-response study since they are more sensitive to pick up the effective doses. We also demonstrate our new approach with application to a pre-clinical dose-response data from Ruberg (1995).
|
