Publication

A MIXTURE LIKELIHOOD APPROACH FOR GENERALIZED LINEAR-MODELS

WEDEL, M. & DESARBO, WS., 1995, In : Journal of Classification. 12, 1, p. 21-55 35 p.

Research output: Contribution to journalArticleAcademicpeer-review

  • M WEDEL
  • WS DESARBO

A mixture model approach is developed that simultaneously estimates the posterior membership probabilities of observations to a number of unobservable groups or latent classes, and the parameters of a generalized linear model which relates the observations, distributed according to some member of the exponential family, to a set of specified covariates within each Class. We demonstrate how this approach handles many of the existing latent class regression procedures as special cases, as well as a host of other parametric specifications in the exponential family heretofore not mentioned in the latent class literature. As such we generalize the McCullagh and Nelder approach to a latent class framework. The parameters are estimated using maximum likelihood, and an EM algorithm for estimation is provided. A Monte Carlo study of the performance of the algorithm for several distributions is provided, and the model is illustrated in two empirical applications.

Original languageEnglish
Pages (from-to)21-55
Number of pages35
JournalJournal of Classification
Volume12
Issue number1
Publication statusPublished - 1995

    Keywords

  • MIXTURE MODELS, GENERALIZED LINEAR MODELS, EM ALGORITHM, MAXIMUM LIKELIHOOD ESTIMATION, MAXIMUM-LIKELIHOOD, EM ALGORITHM, NORMAL-DISTRIBUTIONS, INFORMATION CRITERION, SWITCHING REGRESSIONS, CONVERGENCE, METHODOLOGY, PARAMETERS, SELECTION, CONSUMER

ID: 6426489