Bayesian Dynamic Clustering Factor Models

TL;DR

This paper introduces Bayesian Dynamic Clustering Factor Models that integrate factor analysis with hidden Markov models to analyze multivariate longitudinal data, enabling simultaneous dimension reduction, clustering, and dynamic transition estimation.

Contribution

The paper presents a novel Bayesian framework combining factor models and hidden Markov models for dynamic clustering of multivariate longitudinal data.

Findings

Effective parameter estimation demonstrated on simulated data

Accurate clustering of subjects achieved in simulations

Utility shown through opioid use disorder dataset analysis

Abstract

We propose novel Bayesian Dynamic Clustering Factor Models (BDCFM) for the analysis of multivariate longitudinal data. BDCFM combines factor models with hidden Markov models to concomitantly perform dimension reduction, clustering, and estimation of the dynamic transitions of subjects through clusters. We develop an efficient Gibbs sampler for exploration of the posterior distribution. An analysis of a simulated dataset shows that our inferential approach works well both at parameter estimation and clustering of subjects. Finally, we illustrate the utility of our BDCFM with an analysis of a dataset on opioid use disorder.