Enregistré dans:
Détails bibliographiques
Auteurs principaux: Roy, Asmita, Zhou, Huijuan, Zhao, Ni, Zhang, Xianyang
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2411.10648
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912126011965440
author Roy, Asmita
Zhou, Huijuan
Zhao, Ni
Zhang, Xianyang
author_facet Roy, Asmita
Zhou, Huijuan
Zhao, Ni
Zhang, Xianyang
contents Testing for mediation effect poses a challenge since the null hypothesis (i.e., the absence of mediation effects) is composite, making most existing mediation tests quite conservative and often underpowered. In this work, we propose a subsampling-based procedure to construct a test statistic whose asymptotic null distribution is pivotal and remains the same regardless of the three null cases encountered in mediation analysis. The method, when combined with the popular Sobel test, leads to an accurate size control under the null. We further introduce a Cauchy combination test to construct p-values from different subsample splits, which reduces variability in the testing results and increases detection power. Through numerical studies, our approach has demonstrated a more accurate size and higher detection power than the competing classical and contemporary methods.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10648
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Subsampling-based Tests in Mediation Analysis
Roy, Asmita
Zhou, Huijuan
Zhao, Ni
Zhang, Xianyang
Methodology
Testing for mediation effect poses a challenge since the null hypothesis (i.e., the absence of mediation effects) is composite, making most existing mediation tests quite conservative and often underpowered. In this work, we propose a subsampling-based procedure to construct a test statistic whose asymptotic null distribution is pivotal and remains the same regardless of the three null cases encountered in mediation analysis. The method, when combined with the popular Sobel test, leads to an accurate size control under the null. We further introduce a Cauchy combination test to construct p-values from different subsample splits, which reduces variability in the testing results and increases detection power. Through numerical studies, our approach has demonstrated a more accurate size and higher detection power than the competing classical and contemporary methods.
title Subsampling-based Tests in Mediation Analysis
topic Methodology
url https://arxiv.org/abs/2411.10648