Borrowing strengh in hierarchical Bayes: posterior concentration of the Dirichlet base measure

DOI10.3150/15-BEJ703zbMATH Open1360.62103arXiv1301.0802MaRDI QIDQ282547FDOQ282547

Authors: Xuanlong Nguyen 

Publication date: 12 May 2016

Published in: Bernoulli (Search for Journal in Brave)

Abstract: This paper studies posterior concentration behavior of the base probability measure of a Dirichlet measure, given observations associated with the sampled Dirichlet processes, as the number of observations tends to infinity. The base measure itself is endowed with another Dirichlet prior, a construction known as the hierarchical Dirichlet processes (Teh et al. [J. Amer. Statist. Assoc. 101 (2006) 1566-1581]). Convergence rates are established in transportation distances (i.e., Wasserstein metrics) under various conditions on the geometry of the support of the true base measure. As a consequence of the theory, we demonstrate the benefit of "borrowing strength" in the inference of multiple groups of data - a powerful insight often invoked to motivate hierarchical modeling. In certain settings, the gain in efficiency due to the latent hierarchy can be dramatic, improving from a standard nonparametric rate to a parametric rate of convergence. Tools developed include transportation distances for nonparametric Bayesian hierarchies of random measures, the existence of tests for Dirichlet measures, and geometric properties of the support of Dirichlet measures.

Full work available at URL: https://arxiv.org/abs/1301.0802

Recommendations

• Borrowing strength and borrowing index for Bayesian hierarchical models
• Posterior concentration rates for empirical Bayes procedures with applications to Dirichlet process mixtures
• scientific article; zbMATH DE number 922042
• Posterior consistency of Dirichlet mixtures in density estimation
• Hierarchical Bayes, maximum a posteriori estimators, and minimax concave penalized likelihood estimation
• Posterior convergence rates for Dirichlet mixtures of beta densities
• Posterior consistency of Dirichlet mixtures for estimating a transition density
• Posterior convergence rates of Dirichlet mixtures at smooth densities

zbMATH Keywords

posterior concentrationWasserstein metricsrandom measuresBayesian asymptoticsDirichlet processesgeometry of supporttransportation distances

Mathematics Subject Classification ID

Convergence of probability measures (60B10) Bayesian inference (62F15) Asymptotic properties of nonparametric inference (62G20) Inequalities; stochastic orderings (60E15) Random measures (60G57)

Cites Work

• Asymptotic behaviour of the posterior distribution in overfitted mixture models
• Weak convergence and empirical processes. With applications to statistics
• Title not available (Why is that?)
• A Bayesian analysis of some nonparametric problems
• Optimal Transport
• Convergence rates of posterior distributions.
• Optimal Rates of Convergence for Deconvolving a Density
• On the optimal rates of convergence for nonparametric deconvolution problems
• Statistical decision theory and Bayesian analysis. 2nd ed
• Hierarchical Dirichlet Processes
• Title not available (Why is that?)
• Bayesian Nonparametrics
• Ferguson distributions via Polya urn schemes
• Convergence of latent mixing measures in finite and infinite mixture models
• Fourier methods for estimating mixing densities and distributions
• Rates of convergence of posterior distributions.
• Bayesian nonparametrics
• Rates of contraction of posterior distributions based on Gaussian process priors
• New approaches to Bayesian consistency
• The consistency of posterior distributions in nonparametric problems
• Title not available (Why is that?)
• The tails of probabilities chosen from a Dirichlet prior
• Probability inequalities for likelihood ratios and convergence rates of sieve MLEs
• Posterior contraction of the population polytope in finite admixture models
• Contributions to the theory of Dirichlet processes
• Rates of contraction for posterior distributions in \(L^{r}\)-metrics, \(1 \leq r \leq \infty\)
• About the posterior distribution in hidden Markov models with unknown number of states
• Dimension functions of Cantor sets
• The local geometry of finite mixtures
• On rates of convergence for posterior distributions in infinite-dimensional models

Cited In (10)

• A Wasserstein Index of Dependence for Random Measures
• Measuring dependence in the Wasserstein distance for Bayesian nonparametric models
• Bayesian prediction with multiple-samples information
• Distribution theory for hierarchical processes
• Posterior concentration and fast convergence rates for generalized Bayesian learning
• Hierarchical Dirichlet process and relative entropy
• Title not available (Why is that?)
• Convergence rates of latent topic models under relaxed identifiability conditions
• Convergence of de Finetti's mixing measure in latent structure models for observed exchangeable sequences
• Hierarchical species sampling models

This page was built for publication: Borrowing strengh in hierarchical Bayes: posterior concentration of the Dirichlet base measure

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q282547)