Equivalence of several methods for efficient best subsets selection in generalized linear models

Borko D. Jovanovic*, David W. Hosmer, John P. Buonaccorsi

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

In the recent past, five methods for reducing computational intensity in best subset selection for Generalized Linear Models (GLM) have been proposed. We review these methods and explicitly show their mutual equivalence. Further, we show how the existing linear regression software can be used for such efficient best subset selection. Using the summary results presented in this paper, efficient best subset selection can easily be made available for all nonlinear GLM already present in statistical packages. This is of special importance for computing environments where computational efficiency has a high priority.

Original languageEnglish (US)
Pages (from-to)59-64
Number of pages6
JournalComputational Statistics and Data Analysis
Volume20
Issue number1
DOIs
StatePublished - Jul 1995
Externally publishedYes

Keywords

  • Linearization
  • Logistic regression
  • Mallows' C
  • Poisson regression
  • Pseudo data
  • Survival models

ASJC Scopus subject areas

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Equivalence of several methods for efficient best subsets selection in generalized linear models'. Together they form a unique fingerprint.

Cite this