|
Activity Number:
|
240
- Topics in Multiplicity and Control of False Discovery Rate
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 29, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #306751
|
Presentation
|
|
Title:
|
An Extended Simes Test Procedure for Multiple Testing
|
|
Author(s):
|
Matthew Hudson* and Dr. Joshua Naranjo and Dror Rom
|
|
Companies:
|
Prosoft Clinical and Western Michigan University and Prosoft Clinical
|
|
Keywords:
|
Simes;
strong control;
family-wise error rate;
Multiple Testing
|
|
Abstract:
|
Simes's (1986) improved Bonferroni test has become the foundation of many recent p-value based multiple testing procedures. Simes's test can be applied only as a global test. Hochberg (1988) and Hommel (1988) have extended Simes's test for making inferences on individual hypotheses. However, these procedures and others based on Simes's test are conservative as they may not be able to reject any of the individual hypotheses when Simes's test rejects the global null hypothesis. We devise a procedure for testing a family of hypotheses based on Simes test that can reject at least one hypothesis whenever Simes's test rejects the global null hypothesis. We show that this procedure has strong control of the family-wise error rate under a variety of distributions. We investigate the type-1 error and power for independent as well as correlated statistics in comparison with other Simes based procedures.
|
Authors who are presenting talks have a * after their name.
