A Variable Selection Criterion in the Linear Programming Approaches to Discriminant Analysis

Ravinder Nath, Thomas W. Jones

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

There are numerous variable selection rules in classical discriminant analysis. These rules enable a researcher to distinguish significant variables from nonsignificant ones and thus provide a parsimonious classification model based solely on significant variables. Prominent among such rules are the forward and backward stepwise variable selection criteria employed in statistical software packages such as Statistical Package for the Social Sciences and BMDP Statistical Software. No such criterion currently exists for linear programming (LP) approaches to discriminant analysis. In this paper, a criterion is developed to distinguish significant from nonsignificant variables for use in LP models. This criterion is based on the “jackknife” methodology. Examples are presented to illustrate implementation of the proposed criterion.

Original languageEnglish (US)
Pages (from-to)554-563
Number of pages10
JournalDecision Sciences
Volume19
Issue number3
DOIs
StatePublished - 1988
Externally publishedYes

Fingerprint

Discriminant analysis
Linear programming
Social sciences
Software packages
Selection criteria
Variable selection
Software

All Science Journal Classification (ASJC) codes

  • Business, Management and Accounting(all)
  • Strategy and Management
  • Information Systems and Management
  • Management of Technology and Innovation

Cite this

A Variable Selection Criterion in the Linear Programming Approaches to Discriminant Analysis. / Nath, Ravinder; Jones, Thomas W.

In: Decision Sciences, Vol. 19, No. 3, 1988, p. 554-563.

Research output: Contribution to journalArticle

@article{11eb0a9c97014d8dbf8229122a312cbf,
title = "A Variable Selection Criterion in the Linear Programming Approaches to Discriminant Analysis",
abstract = "There are numerous variable selection rules in classical discriminant analysis. These rules enable a researcher to distinguish significant variables from nonsignificant ones and thus provide a parsimonious classification model based solely on significant variables. Prominent among such rules are the forward and backward stepwise variable selection criteria employed in statistical software packages such as Statistical Package for the Social Sciences and BMDP Statistical Software. No such criterion currently exists for linear programming (LP) approaches to discriminant analysis. In this paper, a criterion is developed to distinguish significant from nonsignificant variables for use in LP models. This criterion is based on the “jackknife” methodology. Examples are presented to illustrate implementation of the proposed criterion.",
author = "Ravinder Nath and Jones, {Thomas W.}",
year = "1988",
doi = "10.1111/j.1540-5915.1988.tb00286.x",
language = "English (US)",
volume = "19",
pages = "554--563",
journal = "Decision Sciences",
issn = "0011-7315",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - A Variable Selection Criterion in the Linear Programming Approaches to Discriminant Analysis

AU - Nath, Ravinder

AU - Jones, Thomas W.

PY - 1988

Y1 - 1988

N2 - There are numerous variable selection rules in classical discriminant analysis. These rules enable a researcher to distinguish significant variables from nonsignificant ones and thus provide a parsimonious classification model based solely on significant variables. Prominent among such rules are the forward and backward stepwise variable selection criteria employed in statistical software packages such as Statistical Package for the Social Sciences and BMDP Statistical Software. No such criterion currently exists for linear programming (LP) approaches to discriminant analysis. In this paper, a criterion is developed to distinguish significant from nonsignificant variables for use in LP models. This criterion is based on the “jackknife” methodology. Examples are presented to illustrate implementation of the proposed criterion.

AB - There are numerous variable selection rules in classical discriminant analysis. These rules enable a researcher to distinguish significant variables from nonsignificant ones and thus provide a parsimonious classification model based solely on significant variables. Prominent among such rules are the forward and backward stepwise variable selection criteria employed in statistical software packages such as Statistical Package for the Social Sciences and BMDP Statistical Software. No such criterion currently exists for linear programming (LP) approaches to discriminant analysis. In this paper, a criterion is developed to distinguish significant from nonsignificant variables for use in LP models. This criterion is based on the “jackknife” methodology. Examples are presented to illustrate implementation of the proposed criterion.

UR - http://www.scopus.com/inward/record.url?scp=84985817878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84985817878&partnerID=8YFLogxK

U2 - 10.1111/j.1540-5915.1988.tb00286.x

DO - 10.1111/j.1540-5915.1988.tb00286.x

M3 - Article

VL - 19

SP - 554

EP - 563

JO - Decision Sciences

JF - Decision Sciences

SN - 0011-7315

IS - 3

ER -