Best-subset model selection based on multitudinal assessments of likelihood improvements

Knute D Carter; Joseph E Cavanaugh

doi:10.1080/02664763.2019.1645097

Back

Best-subset model selection based on multitudinal assessments of likelihood improvements

Journal article

Open access

Peer reviewed

Best-subset model selection based on multitudinal assessments of likelihood improvements

Knute D Carter and Joseph E Cavanaugh

Journal of applied statistics, Vol.47(13-15), pp.2384-2420

11/17/2020

DOI: 10.1080/02664763.2019.1645097

PMCID: PMC9041924

PMID: 35707408

Files and links (1)

url

https://doi.org/10.1080/02664763.2019.1645097View

Published (Version of record) Open Access

Abstract

A common model selection approach is to select the best model, according to some criterion, from among the collection of models defined by all possible subsets of the explanatory variables. Identifying an optimal subset has proven to be a challenging problem, both statistically and computationally. Our model selection procedure allows the researcher to nominate, a priori, the probability at which models containing false or spurious variables will be selected from among all possible subsets. The procedure determines whether inclusion of each candidate variable results in a sufficiently improved fitting term - and is hence named the SIFT procedure. Two variants are proposed: a naive method based on a set of restrictive assumptions and an empirical permutation-based method. Properties of these methods are investigated within the standard linear modeling framework and performance is evaluated against other model selection techniques. The SIFT procedure behaves as designed - asymptotically selecting variables that characterize the underlying data generating mechanism, while limiting selection of spurious variables to the desired level. The SIFT methodology offers researchers a promising new approach to model selection, providing the ability to control the probability of selecting a model that includes spurious variables to a level based on the context of the application.

Akaike information criterion

Bayesian information criterion

likelihood ratio

linear models

regression

variable selection

Details

Title: Subtitle: Best-subset model selection based on multitudinal assessments of likelihood improvements
Creators: Knute D Carter - Department of Biostatistics, University of Iowa
Joseph E Cavanaugh - Department of Biostatistics, University of Iowa
Resource Type: Journal article
Publication Details: Journal of applied statistics, Vol.47(13-15), pp.2384-2420
DOI: 10.1080/02664763.2019.1645097
PMID: 35707408
PMCID: PMC9041924
NLM abbreviation: J Appl Stat
ISSN: 0266-4763
eISSN: 1360-0532
Publisher: Taylor & Francis
Language: English
Date published: 11/17/2020
Academic Unit: Statistics and Actuarial Science; Biostatistics; Injury Prevention Research Center
Record Identifier: 9984214694802771

Metrics

23 Record Views

2 Times Cited - Web of Science