Unified conformalized multiple testing with full data efficiency
Authors
Research Topics
Paper Information
-
Journal:
Journal of the Royal Statistical Society Series B -
DOI:
10.1093/jrsssb/qkag132 -
Published:
October 16, 2026 -
Added to Tracker:
Oct 17, 2026
Abstract
Abstract Conformalized multiple testing offers a model-free way to control predictive uncertainty in decision-making. Existing methods typically use only part of the available data to build score functions tailored to specific settings. We propose a unified framework that puts data utilization at the centre: it uses all available data—null, alternative, and unlabelled—to construct scores and calibrate p-values through a full permutation strategy. This unified use of all available data significantly improves power by enhancing nonconformity score quality and maximizing calibration set size while rigorously controlling the false discovery rate. Crucially, our framework provides a systematic design principle for conformal testing and enables automatic selection of the best conformal procedure among candidates without extra data splitting. Extensive numerical experiments demonstrate that our enhanced methods deliver superior efficiency and adaptability across diverse scenarios.
Author Details
Changliang Zou
AuthorHaojie Ren
AuthorYuyang Huo
AuthorXiaoyang Wu
AuthorResearch Topics & Keywords
Hypothesis Testing
Research AreaCitation Information
APA Format
Changliang Zou
,
Haojie Ren
,
Yuyang Huo
&
Xiaoyang Wu
(2026)
.
Unified conformalized multiple testing with full data efficiency.
Journal of the Royal Statistical Society Series B
, 10.1093/jrsssb/qkag132.
BibTeX Format
@article{paper1749,
title = { Unified conformalized multiple testing with full data efficiency },
author = {
Changliang Zou
and Haojie Ren
and Yuyang Huo
and Xiaoyang Wu
},
journal = { Journal of the Royal Statistical Society Series B },
year = { 2026 },
doi = { 10.1093/jrsssb/qkag132 },
url = { https://doi.org/10.1093/jrsssb/qkag132 }
}