A variable selection technique in discriminant analysis with application in marketing data

A. K. Gupta, T. P. Logan, Jie Chen

Research output: Contribution to journalArticle

4 Scopus citations

Abstract

In this paper, we have developed a likelihood ratio factorization technique for variable selection for multiple observations model. The asymptotic distribution of the selection criterion is also given. This technique has been applied to a marketing data set for illustration of the technique developed.

Original languageEnglish (US)
Pages (from-to)187-199
Number of pages13
JournalJournal of Statistical Computation and Simulation
Volume63
Issue number2
DOIs
Publication statusPublished - Jan 1 1999
Externally publishedYes

    Fingerprint

Keywords

  • Hierarchical model
  • Likelihood ratio factorization technique
  • Multiple observations
  • Multivariate normal distribution

ASJC Scopus subject areas

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this