|
Activity Number:
|
276
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #302174 |
|
Title:
|
Power and Type I Error Comparisons for Seven Multiplicity Adjustment Methods
|
|
Author(s):
|
Jin Xu*+ and Li David and Kenneth Liu and Ivan S.F. Chan and Jie Chen
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc. and Merck & Co., Inc. and Merck Research Laboratories and Merck & Co., Inc.
|
|
Address:
|
351 N. Sumneytown Pike, North Wales, PA, 19454,
|
|
Keywords:
|
Multiplicity adjustment ; Power ; Type I error
|
|
Abstract:
|
Many new multiplicity adjustment procedures have been proposed in the past two decades including methods proposed by Holm, Hochberg, Hommel, Rom, Dunnet and Tamhane, and step-up bootstrap and permutation methods. The last two methods do not require any distribution assumptions. Most statisticians realize the necessity to adjust for multiple hypothesis tests in a clinical trial. The question is which adjustment method ensures control of the Type I error and maximizes the power of the clinical trial. We will expand the research by Dunnet and Tahmane (1993) by evaluating the power of the seven widely used multiplicity adjustment methods listed above under various data structures, including negatively correlated data. We will also complement their research by evaluating the Type I error of these methods for both the one-sided or two-sided testing scenarios.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
