A Hierarchical Model for Time Dependent Multivariate Longitudinal Data
Recently, the use of finite mixture models to cluster three-way data sets has become popular. A natural extension of mixture models to model time dependent data is represented by Hidden Markov models (HMMs) (Cappé et al. 2005); thus, a direct generalization in the finite mixture context for solving the problem of mixing in the time dimension may be given adapting HMMs to three way data clustering. We discuss the issue of longitudinal multivariate data allowing for both time and local dependence.
KeywordsMixture Model Hide Markov Model Latent Class Finite Mixture Finite Mixture Model
- Cappé, O., Moulines, E., & Rydén, T. (2005). Inference in hidden Markov models. Springer series in statistics. Berlin: Springer.Google Scholar
- Everitt, B. S. (1993). Cluster analysis. London: Edward Arnold.Google Scholar
- McLachlan, G. J., & Peel, D. (2000b). Mixture of factor analyzers. In P. Langley (Ed.), Proceedings of the seventeenth international conference on Machine Learnings. San Francisco: Morgan Kauffmann.Google Scholar
- Pearl, J. (1988). Probabilistic reasoning in intelligent systems: Networks of plausible inference. San Matteo, CA: Morgan Kauffmann.Google Scholar
- Vermunt, J. K., Tran, B., & Magidson, J. (2008). Latent class models in longitudinal research. In S. Menard (Ed.), Handbook of longitudinal research: Design, measurement, and analysis (pp. 373–385). Burlington, MA: Elsevier.Google Scholar