Local False Discovery Rate Estimation with Competition-Based Procedures for Variable Selection

TL;DR

This paper introduces TDfdr, a novel method for estimating local false discovery rates using competition-based procedures like knockoff filters, which does not rely on p-values or known null distributions, improving accuracy and applicability.

Contribution

The paper proposes TDfdr, a new approach for local FDR estimation that leverages competition-based procedures, offering a p-value free and more reliable alternative.

Findings

TDfdr accurately estimates FDR in simulations.

TDfdr improves variable selection power in biological datasets.

Method is robust without needing null distribution knowledge.

Abstract

Multiple hypothesis testing has been widely applied to problems dealing with high-dimensional data, e.g., selecting significant variables and controlling the selection error rate. The most prevailing measure of error rate used in the multiple hypothesis testing is the false discovery rate (FDR). In recent years, local false discovery rate (fdr) has drawn much attention, due to its advantage of accessing the confidence of individual hypothesis. However, most methods estimate fdr through p-values or statistics with known null distributions, which are sometimes not available or reliable. Adopting the innovative methodology of competition-based procedures, e.g., knockoff filter, this paper proposes a new approach, named TDfdr, to local false discovery rate estimation, which is free of the p-values or known null distributions. Simulation results demonstrate that TDfdr can accurately estimate the fdr with two competition-based procedures. In real data analysis, the power of TDfdr on variable selection is verified on two biological datasets.