A0590
Title: Tau-censored weighted Benjamini-Hochberg procedures under independence
Authors: Huijuan Zhou - Shanghai University of Finance and Economics (China) [presenting]
Abstract: In the field of multiple-hypothesis testing, auxiliary information can be leveraged to enhance the efficiency of test procedures. A common way to make use of auxiliary information is by weighting p-values. However, when the weights are learned from data, controlling the finite-sample false discovery rate (FDR) becomes challenging, and most existing weighted procedures only guarantee FDR control in an asymptotic limit. In this article, two methods are introduced for constructing data-driven weights for tau-censored weighted Benjamini-Hochberg procedures under independence. They provide new insight into masking p-values to prevent overfitting in multiple tests. The first method utilizes a leave-one-out technique, where all but one of the p-values are used to learn a weight for each p-value. This technique masks the information of a p-value in its weight by calculating the infimum of the weight with respect to the p-value. The second method uses partial information from each p-value to construct weights and utilizes the conditional distributions of the null p-values to establish FDR control. Additionally, two methods are proposed for estimating the null proportion, and how to integrate null-proportion adaptivity into the proposed weights is demonstrated to improve power.
