Activity Number:
|
286
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
Sponsor:
|
ENAR
|
Abstract - #310196 |
Title:
|
Bayesian Sample Size Determination Under Hypothesis Tests
|
Author(s):
|
Gary Cutter*+ and Xiao Zhang
|
Companies:
|
The University of Alabama at Birmingham and The University of Alabama at Birmingham
|
Address:
|
1665 University BLVD Suite 414, Birmingham, AL, 35294,
|
Keywords:
|
Bayes factor ; Hypothesis test; ; Conjugate distribution ; Simulation ; Sensitivity
|
Abstract:
|
We develop a Bayesian approach for calculating sample sizes for clinical trials using the framework of hypothesis tests. The hypotheses we consider are generic and involve specifying conjugate distributions for treatment effects or group difference, and variances of the data. We select sample sizes using the Bayes factor and the averaged type I error and type II error defined by Weiss [1]. Utilizing conjugate distributions, the hypothesis allows the uncertainty inherent in both treatment effects and variances in the estimates from the data. Therefore our approach for calculating sample sizes permits informative prior information for unknown quantities through the hypothesis specification. We investigate the sensitivity of calculated sample sizes to the prior information through simulation. The illustrated example is from an MS clinical trial and the computation is done through WINBUGS1.4
|