Bayesian model averaging in model-based clustering and density estimation

2015/06/30 by Russell, Niamh, Murphy, Thomas Brendan, Raftery, Adrian E
#Computation (stat.CO) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1506.09035

Abstract

We propose Bayesian model averaging (BMA) as a method for postprocessing the results of model-based clustering. Given a number of competing models, appropriate model summaries are averaged, using the posterior model probabilities, instead of being taken from a single "best" model. We demonstrate the use of BMA in model-based clustering for a number of datasets. We show that BMA provides a useful summary of the clustering of observations while taking model uncertainty into account. Further, we show that BMA in conjunction with model-based clustering gives a competitive method for density estimation in a multivariate setting. Applying BMA in the model-based context is fast and can give enhanced modeling performance.

Related

• clusterBMA: Bayesian model averaging for clustering
• A Conceptual Introduction to Bayesian Model Averaging
• Learning All Credible Bayesian Network Structures for Model Averaging
• Bayesian model averaging via mixture model estimation
• Improved model-based clustering performance using Bayesian initialization averaging