Multiple hypothesis testing for variable selection

Rohart, Florian (2016) Multiple hypothesis testing for variable selection. Australian and New Zealand Journal of Statistics, 58 2: 245-267. doi:10.1111/anzs.12157


Author Rohart, Florian
Title Multiple hypothesis testing for variable selection
Journal name Australian and New Zealand Journal of Statistics   Check publisher's open access policy
ISSN 1467-842X
1369-1473
Publication date 2016-06-01
Year available 2016
Sub-type Article (original research)
DOI 10.1111/anzs.12157
Open Access Status Not yet assessed
Volume 58
Issue 2
Start page 245
End page 267
Total pages 23
Place of publication Richmond, VIC, Australia
Publisher Wiley-Blackwell Publishing Asia
Language eng
Abstract We propose two new procedures based on multiple hypothesis testing for correct support estimation in high-dimensional sparse linear models. We conclusively prove that both procedures are powerful and do not require the sample size to be large. The first procedure tackles the atypical setting of ordered variable selection through an extension of a testing procedure previously developed in the context of a linear hypothesis. The second procedure is the main contribution of this paper. It enables data analysts to perform support estimation in the general high-dimensional framework of non-ordered variable selection. A thorough simulation study and applications to real datasets using the R package mht shows that our non-ordered variable procedure produces excellent results in terms of correct support estimation as well as in terms of mean square errors and false discovery rate, when compared to common methods such as the Lasso, the SCAD penalty, forward regression or the false discovery rate procedure (FDR).
Keyword High-dimension
Linear model
Support estimation
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status UQ

Document type: Journal Article
Sub-type: Article (original research)
Collections: HERDC Pre-Audit
UQ Diamantina Institute Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 3 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 3 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Sun, 14 Aug 2016, 10:21:30 EST by System User on behalf of Learning and Research Services (UQ Library)