2020/03/12 by Minwoo Chae, Chae, Minwoo, Pierpaolo De Blasi +3
Computer Science · Mathematics · #62F15 #62G07 #62G20 #Applied mathematics #Bayesian Methods and Mixture Models #Bayesian probability #Consistency (knowledge bases) #Convergence (economics) #Dirichlet distribution #Dirichlet process #Discrete mathematics #Distribution (mathematics) #FOS: Mathematics #Kullback–Leibler divergence #Mathematical analysis #Mathematics #Metric (unit) #Moment (physics) #Point processes and geometric inequalities #Posterior probability #Real line #Statistical Methods and Inference #Statistics #Statistics Theory (math.ST) #Wasserstein metric #math.ST #msc:62F15 #msc:62G07 #msc:62G20 #stat.TH
paper · pdf · doi:10.48550/arxiv.2003.05599
43pages, 4 figures
openalex publication_date 2020/03/12 · arxiv created 2021/06/30 · arxiv updated 2021/07/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
In this paper, we use the class of Wasserstein metrics to study asymptotic properties of posterior distributions. Our first goal is to provide sufficient conditions for posterior consistency. In addition to the well-known Schwartz's Kullback--Leibler condition on the prior, the true distribution and most probability measures in the support of the prior are required to possess moments up to an order which is determined by the order of the Wasserstein metric. We further investigate convergence rates of the posterior distributions for which we need stronger moment conditions. The required tail conditions are sharp in the sense that the posterior distribution may be inconsistent or contract slowly to the true distribution without these conditions. Our study involves techniques that build on recent advances on Wasserstein convergence of empirical measures. We apply the results to density estimation with a Dirichlet process mixture prior and conduct a simulation study for further illustration.