Abstract
This paper is about fitting multivariate normal mixture distributions subject to structural equation modeling. The general model comprises common factor and structural regression models. The introduction of covariance and mean structure models reduces the number of parameters to be estimated in fitting the mixture and enables one to investigate a variety of substantive hypotheses concerning the differences between the components in the mixture. Within the general model, individual parameters can be subjected to equality, nonlinear and simple bounds constraints. Confidence intervals are based on the inverse of the Hessian and on the likelihood profile. Several illustrations are given and results of a simulation study concerning the confidence intervals are reported.
Similar content being viewed by others
References
Agha, M. & Branker, D. S. (1997). Maximum Likelihood estimations and goodness of fit tests for mixtures of distributions (AS 317).Applied Statistics, 46, 399–407.
Aitchison, J., & Silvey, S. D. (1958). Maximum-likelihood estimation of parameters subject to restraints.Annals of Mathematical Statistics, 29, 813–828.
Arminger, G., & Stein, P. (1997). Finite mixtures of covariance structure models with regressions: likelihood, function, minimum distance estimation, fit indices, and a complex example.Submitted.
Azzalini, A. (1996).Statistical inference based on the likelihood. London: Chapman and Hall.
Blåfield, E. (1980).Clustering of observations from finite mixtures with structural information (Uyväskylä studies in computer science, economics and statistics, No. 2). Jyväskylä, Finland: Jyväskylä University.
Dolan, C. V., & Molenaar, P. C. M. (1991). A comparison of 4 methods of calculating standard errors of maximum likelihood estimates in the analysis of covariance structures.British Journal of Mathematical and Statistical Psychology, 44, 359–368.
Everitt, B. S., & Hand, D. J. (1981).Finite mixture distributions. London: Champan and Hall.
Farmacotherapeutisch Kompas (1994).Farmacotherapeutisch kompas: medisch farmaceutisch voorlichting [Pharmacotherapeutic guide: medical pharmaceutical information]. Amstelveen: Ziekenfonsraad.
Feng, Z. D. & McCulloch, C. E. (1996). Using bootstrap likelihood ratios in finite mixture models.Journal of the Royal Statistical Society, Series B, 58, 609–617.
Gill, P. E., Murray, W., & Wright, M. H. (1981).Practical Optimization. London: Academic Press.
Goldberg, D. E. (1989).Genetic algorithms in search, optimization and machine learning. Reading: Addison-Wesley.
Graybill, F. A. (1983).Matrices with applications in statistics (2nd ed.). Belmont, CA: Wadsworth.
Hamilton, J. D. (1990). Analysis of time series subject to changes in regime.Journal of Econometrics, 45, 39–70.
Hamilton, J. D. (1991). A quasi-Bayesian approach to estimating parameters for mixtures of normal distributions.Journal of Business and Economic Statistics, 9, 27–39.
Hathaway, R. J. (1985). A constrained formulation of maximum-likelihood estimation for normal mixture distributions.The Annals of Statistics, 13, 795–800.
Hosmer, D. W. (1974). Maximum Likelihood estimates of parameters of a mixture of two regression lines.Communication in Statistics. Theory and Methods, 3, 995–1006.
Jedidi, K., Jagpal, H. S., & DeSarbo, W. S. (1997a). Finite-mixture structural equation models for response-based segmentation and unobserved heterogeneity.Marketing Science, 16, 39–59.
Jedidi, K., Jagpal, H. S., & DeSarbo, W. S. (1976b). STEMM: A general finite mixture structural equation model.Journal of Classification, 14, 23–50.
Jöreskog, K. G. (1970). Estimation and fitting of simplex models.British Journal of Mathematical and Statistical Psychology, 23, 121–145.
Jöreskog, K. G. (1971). Simultaneous factor analysis in several populations.Psychometrika, 57, 409–426.
Jöreskog, K. G. (1977). Structural equation models in the social sciences: Specification, estimation and testing. In P. R. Krishnaiah (Ed.),Applications of Statistics. Amsterdam: North-Holland.
Jöreskog, K. G., & Sörbom, D. (1993).LISREL 8 user's reference and guide. Chicago: Scientific Software International.
McLachlan, G. J. (1987). On bootstrapping the likelihood ratio test statistic for the number of components in a normal mixture.Applied Statistics, 36, 318–324.
Murphy, A. E., & Bolling, D. R. (1967). Testing of a single locus hypothesis where there is incomplete separation of the phenotypes.American Journal of Human Genetics, 19, 322–334.
Numerical Alogorithms Group. (1990).The NAG Fortran Manual, Mark 4. Oxford: Author.
Neale, M. C. (1995).Mx: Statistical Modeling (3rd ed.). Richmond, VA: Medical College of Virginia.
Neale, M. C., & Miller, M. (1997). The use of likelihood-based confidence intervals in genetic models.Behavior Genetics, 27, 113–120.
Piaget, J., & Inhelder, B. (1969).The psychology of the child. New York: Basic Books.
Sörbom, D. (1974). A general method for studying differences in factor means and factor structures between groups.British Journal of Mathematical and Statistical Psychology, 27, 229–239.
Stein, P. (1997).Mischungen von konditionalen Mittlewertund Kovarianzstrukturmodellen mit Anwendungen auf die analyse von Lebensstilen. Unpublished doctoral dissertation, Department of Social Science, Gerhard Mercator University, Disburg, Germany.
Titterington, D. M., Smith, A. F. M., & Makov, U. E. (1985).Statistical analysis of finite mixture distributions. Chicester: John Wiley & Sons.
Van der Maas, H. J. L. (1993).Catastrophe analysis of stagewise cognitive development: models, method and applications (Dissertatie reeks 1993-2). Amsterdam: University of Amsterdam, Psychology Faculty.
van der Maas, H. J. L., & Raijmakers, M. E. J. (1997).Optimizing latent class models by genetic alogorithms (Internal Report). Amsterdam: University of Amsterdam, Developmeental Psychology, Psychology Faculty.
Venzon, D. J., & Moolgavkar, S. H. (1988). A method for computing profile-likelihood-based confidence intervals.Applied Statistics, 37, 87–94.
Wedel, M. (1995). GLIMMIX, A program for mixtures of generalized linear regression models, and its applications in marketing.Kwantitatieve Methoden, 50, 55–70.
Wedel, M., & DeSarbo, W. S. (1994). A review of recent developments in latent class regression models. In R. P. Bagozzi (Ed.),Advanced methods of marketing research (pp. 352–388) Cambridge, MA: Blackwell.
Wedel, M., & DeSarbo, W. S. (1995). A mixture likelihood approach for generalized linear models.Journal of Classification, 12, 21–55.
Wolfe, J. H. (1970). Pattern clustering by multivariate mixture analysis.Multivariate Behavioral Research, 5, 329–350.
Yung, Y. F. (1994).Finite mixtures in confirmatory factor-analytic models. Unpublished doctoral dissertation, University of California, Los Angeles. (Also available as Yung, Y. F. 1995,Finite Mixtures in Confirmatory Factor-Analytic Models (microfilm). Ann Arbor, MI: University Microfilms.
Yung, Y. F. (1997). Finite Mixtures in Confirmatory Factor-Analysis Models.Psychometrika, 62, 297–330.
Author information
Authors and Affiliations
Additional information
We thank Harrie Vorst for his generous support. We thank Yiu-Fai Yung for kindly making available his thesis (Yung, 1994), and a copy of his, at the time of revising the present paper, unpublished Psychometrika article (Yung, 1997). We thank Petra Stein for making available her paper with Gerhard Arminger (Arminger & Stein, 1997). A total of five reviews provided by anonymous referees, and additional comments by the editors resulted in a good number of improvements. Finally, we thank Peter Molenaar for his critical comments.
This paper was first submitted to Psychometrika in May of 1996, after a period of about 6 months in which we developed and tested our FORTRAN routines. In two rounds of highly constructive reviews, two dissertations (Stein, 1997; Yung, 1994), and severl articles were brought to our attention, that were either submitted (Arminger & Stein), or in press (Yung, Jedidi et al.). In the mean time, a number of these papers have appeared (Jedidi et al. 1997a, 1997b), and more are sure to follow (Yund, 1997; Arminger & Stein, 1997). It is clear that the subject of structural equation modeling within normal mixtures has taken off over the past few years (though Blåfield's pioneering thesis appeared in 1980). Although we were unaware of the work that was on-going, or indeed completed, when we embarked on this project, it is with pleasure that we acknowledge the precedence of those, whose work, in press, submitted, or otherwise, has come to our attention sice first submitting this paper to Psychometrika.
Rights and permissions
About this article
Cite this article
Dolan, C.V., van der Maas, H.L.J. Fitting multivariage normal finite mixtures subject to structural equation modeling. Psychometrika 63, 227–253 (1998). https://doi.org/10.1007/BF02294853
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02294853