Consistency of mixture models with a prior on the number of components

DOI10.1515/DEMO-2022-0150zbMATH Open1514.62121arXiv2205.03384OpenAlexW4323780673MaRDI QIDQ6160720FDOQ6160720

Authors: Jeffrey W. Miller 

Publication date: 26 June 2023

Published in: Dependence Modeling (Search for Journal in Brave)

Abstract: This article establishes general conditions for posterior consistency of Bayesian finite mixture models with a prior on the number of components. That is, we provide sufficient conditions under which the posterior concentrates on neighborhoods of the true parameter values when the data are generated from a finite mixture over the assumed family of component distributions. Specifically, we establish almost sure consistency for the number of components, the mixture weights, and the component parameters, up to a permutation of the component labels. The approach taken here is based on Doob's theorem, which has the advantage of holding under extraordinarily general conditions, and the disadvantage of only guaranteeing consistency at a set of parameter values that has probability one under the prior. However, we show that in fact, for commonly used choices of prior, this yields consistency at Lebesgue-almost all parameter values -- which is satisfactory for most practical purposes. We aim to formulate the results in a way that maximizes clarity, generality, and ease of use.

Full work available at URL: https://arxiv.org/abs/2205.03384

Recommendations

• Convergence of posterior distribution in the mixture of regressions
• On Consistency of Nonparametric Normal Mixtures for Bayesian Density Estimation
• Consistency of the MLE under mixture models
• On Consistency of Bayesian Inference with Mixtures of Logistic Regression
• On the posterior distribution of the number of components in a finite mixture

zbMATH Keywords

asymptoticsclusteringBayesian statisticsnonparametric inference

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cites Work

• Practical Bayesian Density Estimation Using Mixtures of Normals
• On the Identifiability of Finite Mixtures
• Real Analysis and Probability
• Fundamentals of nonparametric Bayesian inference
• Identifiability of Finite Mixtures of Elliptical Distributions
• Inconsistency of Pitman-Yor process mixtures for the number of components
• Mixture models with a prior on the number of components
• Robust Bayesian inference via coarsening
• Convergence of latent mixing measures in finite and infinite mixture models
• Gibbs Sampling Methods for Stick-Breaking Priors
• Adaptive Bayesian density estimation with location-scale mixtures
• Title not available (Why is that?)
• Adaptive Bayesian multivariate density estimation with Dirichlet mixtures
• Identifiability of Finite Mixtures
• Concepts of Independence for Proportions with a Generalization of the Dirichlet Distribution
• Title not available (Why is that?)
• Identifiability of mixtures of power-series distributions and related characterizations
• On posterior contraction of parameters and interpretability in Bayesian mixture modeling

Cited In (6)

• On the posterior distribution of the number of components in a finite mixture
• On posterior contraction of parameters and interpretability in Bayesian mixture modeling
• On Consistency of Bayesian Inference with Mixtures of Logistic Regression
• Convergence of posterior distribution in the mixture of regressions
• Inconsistency of Pitman-Yor process mixtures for the number of components
• Bayesian mixture models (in)consistency for the number of clusters

This page was built for publication: Consistency of mixture models with a prior on the number of components

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6160720)