|
Activity Number:
|
403
- SPAAC Poster Competition
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #301779
|
|
|
Title:
|
A Data-Driven Fallback Procedure for Multiple Comparisons
|
|
Author(s):
|
Jared Wolf* and Hong Zhou
|
|
Companies:
|
J.B. Hunt Transport Inc. and Arkansas State University
|
|
Keywords:
|
Multiplicity;
Familywise error rate;
Holm procedure;
Multiple hypotheses testing;
Fallback procedure;
Strong and weak control
|
|
Abstract:
|
The proposed data-driven fallback procedure combines the positive aspects of data-driven multiple hypotheses testing techniques and the positive aspects of those relying on predetermined strategies. This proposed procedure tests hypotheses based on p-value ordering, but the significance level at each sequential step is accumulated in a manner similar to that of the traditional fallback procedure. It has been proven that this proposed procedure strongly controls the familywise error rate. In addition, it has been proven that this procedure is uniformly more powerful than the weighted Holm procedure when more than two hypotheses are tested, and at least as powerful as the weighted Holm procedure for only two hypotheses. A simulation study was conducted to show that the proposed procedure is more powerful than the traditional fallback in most cases.
|
Authors who are presenting talks have a * after their name.
