Overview
Cong Han was a graduate student in the Division of Biostatistics at the University of Minnesota when co-authoring with Bradley P. Carlin the comparative review of MCMC methods for computing Bayes factors (Han-Carlin 2001, JASA 96).
Key Contributions
- Han-Carlin (2001) comparative review — benchmarked six MCMC methods for Bayes factor computation (Carlin-Chib product space, Metropolized CC, reversible jump, partial analytic structure, Chib MLI, Chib-Jeliazkov) across three empirical examples; documented the failure of product-space samplers for large hierarchical models (467 AIDS subjects, 1.5M iterations, no convergence) and identified marginal likelihood and reversible jump methods as the practical workhorses. See Marginal Data Density.
Related