|
Activity Number:
|
177
- Big Data and Computationally Intensive Methods
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 29, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
|
Abstract #307177
|
|
|
Title:
|
A Data-Driven Multiple Testing Procedure
|
|
Author(s):
|
Nasrine Bendjilali* and Boualem Bendjilali and Wei-Min Huang
|
|
Companies:
|
Rowan University and RVCC and Lehigh University
|
|
Keywords:
|
multiple testing procedures;
False Discovery Rate
|
|
Abstract:
|
Benjamini and Hochberg presented a new way to control for the multiplicity problem, which calls for controlling the false discovery rate (FDR). Their procedure (the BH procedure), was shown to control the FDR for any combination of true and false null hypotheses tested. Testing large-scale multiple hypotheses often produce p-values with complex dependency structure. We present a data-driven multiple testing procedure based on BH that incorporates the dependency structure of the p-values and adjust the original BH procedure to provide a tighter control of the false discovery rate (FDR). The proposed procedure is simple to use and shows a significant improvement in power relative to its competitors. Theoretical properties and power performance study will be presented.
|
Authors who are presenting talks have a * after their name.
