Post hoc false positive control for structured hypotheses - Archive ouverte HAL
Journal Articles Scandinavian Journal of Statistics Year : 2020

Post hoc false positive control for structured hypotheses

Abstract

In a high‐dimensional multiple testing framework, we present new confidence bounds on the false positives contained in subsets S of selected null hypotheses. These bounds are post hoc in the sense that the coverage probability holds simultaneously over all S, possibly chosen depending on the data. This article focuses on the common case of structured null hypotheses, for example, along a tree, a hierarchy, or geometrically (spatially or temporally). Following recent advances in post hoc inference, we build confidence bounds for some pre-specified forest‐structured subsets and deduce a bound for any subset S by interpolation. The proposed bounds are shown to improve substantially previous ones when the signal is locally structured. Our findings are supported both by theoretical results and numerical experiments. Moreover, our bounds can be obtained by an algorithm (with complexity bilinear in the sizes of the reference hierarchy and of the selected subset) that is implemented in the open‐source R package sansSouci available from https://github.com/pneuvial/sanssouci, making our approach operational.
Fichier principal
Vignette du fichier
DBNR2020_HAL.pdf (628.98 Ko) Télécharger le fichier
IllustrationBound.pdf (9.38 Ko) Télécharger le fichier
NonposthocCB1.pdf (5.9 Ko) Télécharger le fichier
NonposthocCB2.pdf (5.96 Ko) Télécharger le fichier
Posthocbounds.pdf (6.3 Ko) Télécharger le fichier
all-conf-env_all-alpha_order=p-value_setting=const.pdf (12.57 Ko) Télécharger le fichier
all-conf-env_alpha=0-05_order=p-value_setting=const.pdf (13.68 Ko) Télécharger le fichier
all-conf-env_hybrid_order=p-value_setting=const.pdf (10.38 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01829037 , version 1 (03-07-2018)
hal-01829037 , version 2 (18-09-2018)
hal-01829037 , version 3 (01-07-2020)

Identifiers

Cite

Guillermo Durand, Gilles Blanchard, Pierre Neuvial, Etienne Roquain. Post hoc false positive control for structured hypotheses. Scandinavian Journal of Statistics, 2020, 47 (4), pp.1114-1148. ⟨10.1111/sjos.12453⟩. ⟨hal-01829037v3⟩
431 View
500 Download

Altmetric

Share

More