Response-guided knockoffs for directional FDR control in linear models
Abstract
We consider the problem of feature selection in linear models with finite-sample control of the false discovery rate (FDR). While existing knockoff-based methods control the directional FDR, which penalises incorrect sign estimates, they do not target discoveries in a pre-specified direction, and their knockoff constructions are entirely response-agnostic. We introduce the response-guided knockoff filter, which leverages a noise-perturbed version of the response to guide knockoff construction toward features likely to have the target sign, while provably controlling the directional FDR. The method operates under a weaker sample-size requirement n > p + 2, compared to n ≥ 2p required by existing fixed-X generators. Simulations and HIV drug resistance experiments demonstrate power gains over existing methods.
Related
- Evolving Towards Better Codes: LLM-Guided Search for High-Distance Binary Linear Codes
- Control of 1D first-order linear hyperbolic systems: a brief survey
- Feedback Stabilization for Sampled Linear Systems with Control-linear Noise
- An Optimal Control Approach to Finite-Time Stabilization of Linear Systems
- Linear Response Theory for Jump-Driven Stochastic Systems: Transient Statistics and Escape Dynamics
BibTeX
@misc{freestone2026responseguided,
title = {{Response-guided knockoffs for directional FDR control in linear models}},
author = {Jack Freestone and Garth Tarr and Samuel Muller and Uri Keich},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.29083},
eprint = {2608.29083},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.29083},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Response-guided knockoffs for directional FDR control in linear models
AU - Jack Freestone
AU - Garth Tarr
AU - Samuel Muller
AU - Uri Keich
PY - 2026
DA - 2026/08/29
PB - arXiv
DO - 10.48550/arxiv.2608.29083
UR - https://arxiv.org/abs/2608.29083
AB - We consider the problem of feature selection in linear models with finite-sample control of the false discovery rate (FDR). While existing knockoff-based methods control the directional FDR, which penalises incorrect sign estimates, they do not target discoveries in a pre-specified direction, and their knockoff constructions are entirely response-agnostic. We introduce the response-guided knockoff filter, which leverages a noise-perturbed version of the response to guide knockoff construction toward features likely to have the target sign, while provably controlling the directional FDR. The method operates under a weaker sample-size requirement n > p + 2, compared to n ≥ 2p required by existing fixed-X generators. Simulations and HIV drug resistance experiments demonstrate power gains over existing methods.
ER -