Michael D. Escobar

Overview

Michael D. Escobar is a biostatistician (University of Toronto / Centre for Addiction and Mental Health), known for Bayesian nonparametrics — particularly computational inference for Dirichlet-process mixture models.

Key Contributions / Features

• DP-mixture density estimation (Escobar-West 1995): with West, the Pólya-urn Gibbs sampler for mixtures of Dirichlet processes and the auxiliary-variable update that learns the concentration parameter $\alpha$.
• Earlier work (Escobar 1994) on estimating normal means via Dirichlet-process priors; applications of Bayesian nonparametrics in biostatistics.

Related

• Dirichlet Process Mixture
• Escobar-West (1995) Bayesian Density Estimation and Inference Using Mixtures
• Mike West