A1158
Title: Stabilized eBH: A unified stability approach to false discovery rate control
Authors: Wei Zhong - Xiamen University (China) [presenting]
Jiajun Sun - Xiamen University (China)
Zhanrui Cai - University of Hong Kong (Hong Kong)
Abstract: Stability and reproducibility are essential considerations in various applications of statistical methods to scientific research. False discovery rate (FDR) control methods provide a framework for controlling false signals in scientific discoveries. Many FDR control techniques, such as knockoff and data-splitting approaches, have been successfully developed in multiple testing and regression contexts. However, some of these methods yield unstable results due to the inherent randomness of the algorithms. For instance, different constructions of knockoff copies can lead to different sets of selected variables. To enhance the stability and reproducibility of statistical outcomes, a unified stability approach is proposed for feature selection and multiple testing algorithms with FDR control, named Stabilized eBH. The method aggregates e-values based on rank statistics generated from multiple runs of the base algorithm to construct stabilized e-values, which are then processed using the eBH procedure. This approach not only guarantees FDR control and power performance but also enhances stability. It is adaptable and can be applied to most existing FDR control methods. Moreover, the theoretical properties of the stability method are investigated, including asymptotic FDR control, power, and stability guarantees. Extensive numerical experiments and applications to real datasets demonstrate that the proposed method generally outperforms existing alternatives.
