A discrete modification of the Benjamini-Yekutieli procedure

TL;DR

This paper introduces a modified Benjamini-Yekutieli procedure tailored for discrete test statistics, improving false discovery rate control in dependent data scenarios, with demonstrated benefits in pharmacovigilance and simulations.

Contribution

It proposes a novel modification of the Benjamini-Yekutieli procedure that accounts for discreteness, enhancing its performance under dependence.

Findings

Improved false discovery rate control in discrete settings

Enhanced procedure performance demonstrated in pharmacovigilance data

Simulation results show better power compared to traditional methods

Abstract

The Benjamini-Yekutieli procedure is a multiple testing method that controls the false discovery rate under arbitrary dependence of the $p$-values. A modification of this and related procedures is proposed for the case when the test statistics are discrete. It is shown that taking discreteness into account can improve upon known procedures. The performance of this new procedure is evaluated for pharmacovigilance data and in a simulation study.