Improving knockoffs with conditional calibration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Luo, Yixiang, Fithian, William, Lei, Lihua
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912131224436736
author Luo, Yixiang
Fithian, William
Lei, Lihua
author_facet Luo, Yixiang
Fithian, William
Lei, Lihua
contents The knockoff filter of Barber and Candes (arXiv:1404.5609) is a flexible framework for multiple testing in supervised learning models, based on introducing synthetic predictor variables to control the false discovery rate (FDR). Using the conditional calibration framework of Fithian and Lei (arXiv:2007.10438), we introduce the calibrated knockoff procedure, a method that uniformly improves the power of any fixed-X or model-X knockoff procedure. We show theoretically and empirically that the improvement is especially notable in two contexts where knockoff methods can be nearly powerless: when the rejection set is small, and when the structure of the design matrix in fixed-X knockoffs prevents us from constructing good knockoff variables. In these contexts, calibrated knockoffs even outperform competing FDR-controlling methods like the (dependence-adjusted) procedure Benjamini-Hochberg in many scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2208_09542
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Improving knockoffs with conditional calibration
Luo, Yixiang
Fithian, William
Lei, Lihua
Methodology
62H15 (Primary), 62J15 (Secondary)
The knockoff filter of Barber and Candes (arXiv:1404.5609) is a flexible framework for multiple testing in supervised learning models, based on introducing synthetic predictor variables to control the false discovery rate (FDR). Using the conditional calibration framework of Fithian and Lei (arXiv:2007.10438), we introduce the calibrated knockoff procedure, a method that uniformly improves the power of any fixed-X or model-X knockoff procedure. We show theoretically and empirically that the improvement is especially notable in two contexts where knockoff methods can be nearly powerless: when the rejection set is small, and when the structure of the design matrix in fixed-X knockoffs prevents us from constructing good knockoff variables. In these contexts, calibrated knockoffs even outperform competing FDR-controlling methods like the (dependence-adjusted) procedure Benjamini-Hochberg in many scenarios.
title Improving knockoffs with conditional calibration
topic Methodology
62H15 (Primary), 62J15 (Secondary)
url https://arxiv.org/abs/2208.09542