|
Activity Number:
|
280
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
WNAR
|
| Abstract - #309661 |
|
Title:
|
Optimal Allocation for Sequential Binary Response Experiments
|
|
Author(s):
|
William Rosenberger*+ and Yevgen Tymofyeyev and Feifang Hu
|
|
Companies:
|
George Mason University and Merck & Co., Inc. and University of Virginia
|
|
Address:
|
4400 University Dr MS 4A7, Fairfax, VA, 22030,
|
|
Keywords:
|
adaptive designs ; optimal allocation ; sequential estimation ; binary response
|
|
Abstract:
|
For sequential experiments with K treatments, we establish two formal optimization criteria to find optimal allocation strategies. Both criteria involve the sample sizes on each treatment and a concave noncentrality parameter from a multivariate test. We show these two criteria are equivalent. We apply this result to specific questions: how do we maximize power of a multivariate test of homogeneity with binary response? For fixed power, how do we minimize expected treatment failures? Because the solutions depend on unknown parameters, we describe a response-adaptive randomization procedure which ``targets" the optimal allocation and provides increases in power along the lines of 2--4 percent over complete randomization for equal allocation.
|
