Finite Mixtures of Covariance Structure Models with Regressors
Authors:
- 1 November 1997
- journal article
- Published by SAGE Publications in Sociological Methods & Research
- Vol. 26 (2) , 148-182
Abstract
Models of finite mixtures of normal densities conditional on regressor variables are specified and estimated. The authors consider mixtures of multivariate normals where the expected value for each component depends on nonnormal regressor variables. The expected values and covariance matrices of the mixture components are parameterized using conditional mean- and covariance-structures. The authors discuss the construction of the likelihood function and outline the estimation of parameters. In addition, they define fit indices and discuss aspects of model specification and modification that are specific to mixtures of mean- and covariance-structures. Finally, they give an empirical example in which they analyze the importance of automobiles to individuals depending on the latent constructs individualism and ecology-mindedness. It is shown that the sample under consideration comes from three heterogeneous subpopulations. It is demonstrated that each subpopulation may be characterized by a different mean- and covariance-structure.Keywords
This publication has 9 references indexed in Scilit:
- Mixtures of Conditional Mean- and Covariance-Structure ModelsPsychometrika, 1999
- Finite Mixtures in Confirmatory Factor-Analysis ModelsPsychometrika, 1997
- The distribution of the likelihood ratio for mixtures of densities from the one-parameter exponential familyAnnals of the Institute of Statistical Mathematics, 1994
- A maximum likelihood methodology for clusterwise linear regressionJournal of Classification, 1988
- Estimation and Hypothesis Testing in Finite Mixture ModelsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1985
- Finding the Observed Information Matrix When Using the EM AlgorithmJournal of the Royal Statistical Society Series B: Statistical Methodology, 1982
- Statistical Modelling of Data on Teaching StylesJournal of the Royal Statistical Society. Series A (General), 1981
- Maximum Likelihood from Incomplete Data Via the EM AlgorithmJournal of the Royal Statistical Society Series B: Statistical Methodology, 1977
- Exploratory latent structure analysis using both identifiable and unidentifiable modelsBiometrika, 1974