EUVIMEDSkopri EUVIMED
Crossref

False Discovery Rate Control for Confounder Selection Using Mirror Statistics

Kazuharu Harada iD, Masataka Taguri iD

Statistics in Medicine · 2025

DOI: 10.1002/sim.70116 ↗

At a glance

Journal article in Statistics in Medicine, 2025. Written by Kazuharu Harada and 1 co-authors. Institution: Department of Health Data Science, Tokyo Medical University, Tokyo, Japan. Major MEDLINE topics: Models, Statistical, Observational Studies as Topic. 43 references. Funded by: Japan Society for the Promotion of Science.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

While data‐driven confounder selection requires careful consideration, it is frequently employed in observational studies. Widely recognized criteria for confounder selection include the minimal‐set approach, which involves selecting variables relevant to both treatment and outcome, and the union‐set approach, which involves selecting variables associated with either treatment or outcome. These approaches are often implemented using heuristics and off‐the‐shelf statistical methods, where the degree of uncertainty may not be clear. In this paper, we focus on the false discovery rate (FDR) to measure uncertainty in confounder selection. We define the FDR specific to confounder selection and propose methods based on the mirror statistic, a recently developed approach for FDR control that does not rely on p ‐values. The proposed methods are p ‐value‐free and require only the assumption of some symmetry in the distribution of the mirror statistic. It can be combined with sparse estimation and other methods that involve difficulties in deriving p ‐values. The properties of the proposed methods are investigated through exhaustive numerical experiments. Particularly in high‐dimensional data scenarios, the proposed methods effectively control FDR and perform better than the p ‐value‐based methods.

Subjects

MeSH terms (MEDLINE)

Models, StatisticalObservational Studies as TopicHumansConfounding Factors, EpidemiologicData Interpretation, StatisticalComputer SimulationFalse Positive ReactionsUncertainty

Author keywords

causal inferenceconfounder selectionfalse discovery ratehigh‐dimensional datamirror statistics

Authors and institutions

  • Department of Health Data Science, Tokyo Medical University, Tokyo, Japan.

ORCID profiles: Kazuharu Harada, Masataka Taguri

Funding

  • Japan Society for the Promotion of Science

Cite

APA

Harada, K., & Taguri, M. (2025). False Discovery Rate Control for Confounder Selection Using Mirror Statistics. Statistics in Medicine. https://doi.org/10.1002/sim.70116

Vancouver

Harada K, Taguri M. False Discovery Rate Control for Confounder Selection Using Mirror Statistics. Statistics in Medicine. 2025. doi:10.1002/sim.70116

BibTeX
@article{Harada2025,
  author = {Harada, Kazuharu and Taguri, Masataka},
  title = {False Discovery Rate Control for Confounder Selection Using Mirror Statistics},
  journal = {Statistics in Medicine},
  year = {2025},
  doi = {10.1002/sim.70116}
}
Start for free →