MCMC Inference for Model-based Cluster analysis
A new approach to cluster analysis has been introduced based on parsimonious geometric modelling of the within-group covariance matrices in a mixture of multivariate normal distributions, using Bayesian calculation and the Gibbs sampler. The approach answers many limitations of the Maximum likelihood approach. Here we propose to investigate more general models dealing with the shape and orientation of the clusters using the Gibbs Sampler.
KeywordsBayes Factor Gaussian Mixture Gibbs Sampler
Unable to display preview. Download preview PDF.
- Anderson, T. W. ( 1984, 1958). An Introduction to Multivariate Statistical Analysis. Wiley, New York.Google Scholar
- Dasgupta, A. & Raftery, A. E. (1995). Detecting features in spacial point processes with clutter via Model-based clustering, Technical Report No 295, Department of Statistics, University of Washington.Google Scholar
- Lewis, S. M. & Raftery, A. E. (1997). Estimating Bayes factors via posterior simulation with the Laplace-Metropolis estimator, Journal of the American Statistical Association, to appear. McLachlan, G. & Basford, K. (1988). Mixture Models: Inference and Applications to Clustering. Marcel Dekker, New York.Google Scholar