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

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 su...

Full description

Saved in:
Bibliographic Details
Published in:Computational statistics & data analysis 1995-07, Vol.20 (1), p.59-64
Main Authors: Jovanovic, Borko D., Hosmer, David W., Buonaccorsi, John P.
Format: Article
Language:English
Subjects:
Exact sciences and technology
Linear inference, regression
Linearization
Logistic regression
Mallows' C
Mathematics
Poisson regression
Probability and statistics
Pseudo data
Sciences and techniques of general use
Statistics
Survival models
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:0167-9473
1872-7352
DOI:10.1016/0167-9473(94)00030-M