Controlling the false discovery rate in transformational sparsity: Split Knockoffs
Author:
Cao Yang1,
Sun Xinwei2,
Yao Yuan1
Affiliation:
1. Department of Mathematics, The Hong Kong University of Science and Technology , Clear Water Bay, Kowloon, Hong Kong SAR , People’s Republic of China
2. School of Data Science, Fudan University , Handan Road, Shanghai 200433 , People’s Republic of China
Abstract
Abstract
Controlling the False Discovery Rate (FDR) in a variable selection procedure is critical for reproducible discoveries, and it has been extensively studied in sparse linear models. However, it remains largely open in scenarios where the sparsity constraint is not directly imposed on the parameters but on a linear transformation of the parameters to be estimated. Examples of such scenarios include total variations, wavelet transforms, fused LASSO, and trend filtering. In this paper, we propose a data-adaptive FDR control method, called the Split Knockoff method, for this transformational sparsity setting. The proposed method exploits both variable and data splitting. The linear transformation constraint is relaxed to its Euclidean proximity in a lifted parameter space, which yields an orthogonal design that enables the orthogonal Split Knockoff construction. To overcome the challenge that exchangeability fails due to the heterogeneous noise brought by the transformation, new inverse supermartingale structures are developed via data splitting for provable FDR control without sacrificing power. Simulation experiments demonstrate that the proposed methodology achieves the desired FDR and power. We also provide an application to Alzheimer’s Disease study, where atrophy brain regions and their abnormal connections can be discovered based on a structural Magnetic Resonance Imaging dataset.
Funder
National Natural Science Foundation of China
Research Grants Council Joint Research Scheme
Hong Kong Research Grant Council
State Key Program of National Natural Science Foundation of China
HKRGC Collaborative Research
Publisher
Oxford University Press (OUP)
Subject
Statistics, Probability and Uncertainty,Statistics and Probability
Reference41 articles.
1. A fast diffeomorphic image registration algorithm;Ashburner;Neuroimage,2007
2. Controlling the false discovery rate via Knockoffs;Barber;The Annals of Statistics,2015
3. A Knockoff filter for high-dimensional selective inference;Barber;The Annals of Statistics,2019
4. Robust inference with Knockoffs;Barber;The Annals of Statistics,2020
5. Controlling the false discovery rate: A practical and powerful approach to multiple testing;Benjamini;Journal of the Royal Statistical Society: Series B (Methodological),1995
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献