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 #304771
|
Presentation
|
Title:
|
Incorporating the Sample Correlation Between Two Test Statistics to Adjust the Critical Points for the Control of Type-1 Error
|
Author(s):
|
Dror Rom* and Jaclyn Ashley McTague
|
Companies:
|
Prosoft Clinical and Prosoft Clinical
|
Keywords:
|
Bonferroni;
Multiple Comparisons;
Multiple Endpoints;
Sample Covariance Matrix;
Strong Control of the FWER
|
Abstract:
|
We develop a multiple testing procedure that explicitly incorporates the sample correlation to adjust the critical points for the control of type-1 error rate. For samples from normal random variables, we adjust the critical points using the joint distribution of the sample variance-covariance matrix, and the t-statistics used to test the individual hypotheses. Some analytical results for this procedure are discussed. We also provide a table of critical points for varying degrees of freedom and observed sample correlations which control the type-1 error. The procedure is illustrated with some examples from clinical trials with two endpoints, and with some other multiple testing scenarios. Comparisons are made to other commonly used procedures such as step-up and step-down Bonferroni procedures.
|
Authors who are presenting talks have a * after their name.