Posterior sampling from -approximation of normalized completely random measure mixtures

DOI10.1214/16-EJS1168zbMATH Open1358.62034arXiv1507.04528OpenAlexW2553818082MaRDI QIDQ502819FDOQ502819

Authors: Ilaria Bianchini, Alessandra Guglielmi, Raffaele Argiento 

Publication date: 11 January 2017

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: This paper adopts a Bayesian nonparametric mixture model where the mixing distribution belongs to the wide class of normalized homogeneous completely random measures. We propose a truncation method for the mixing distribution by discarding the weights of the unnormalized measure smaller than a threshold. We prove convergence in law of our approximation, provide some theoretical properties and characterize its posterior distribution so that a blocked Gibbs sampler is devised. The versatility of the approximation is illustrated by two different applications. In the first the normalized Bessel random measure, encompassing the Dirichlet process, is introduced; goodness of fit indexes show its good performances as mixing measure for density estimation. The second describes how to incorporate covariates in the support of the normalized measure, leading to a linear dependent model for regression and clustering.

Full work available at URL: https://arxiv.org/abs/1507.04528

Recommendations

• On the consistency of posterior mixtures and its applications
• Posterior consistency of Dirichlet mixtures in density estimation
• Uniform mixtures via posterior means
• Posterior sampling when the normalizing constant is unknown
• Posterior convergence rates of Dirichlet mixtures at smooth densities

zbMATH Keywords

Dirichlet processBayesian nonparametric mixture modelsblocked Gibbs samplerfinite dimensional approximationnormalized completely random measures

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Random measures (60G57)

Cited In (20)

• MCMC Computations for Bayesian Mixture Models Using Repulsive Point Processes
• Randomized maximum likelihood based posterior sampling
• Hierarchical Normalized Completely Random Measures to Cluster Grouped Data
• Truncated random measures
• Modelling and computation using NCoRM mixtures for density regression
• A blocked Gibbs sampler for NGG-mixture models via a priori truncation
• Importance conditional sampling for Pitman-Yor mixtures
• Capture-Recapture Models with Heterogeneous Temporary Emigration
• Title not available (Why is that?)
• Computational challenges and temporal dependence in Bayesian nonparametric models
• A review of uncertainty quantification for density estimation
• Non-parametric sampling approximation via Voronoi tessellations
• A unified construction for series representations and finite approximations of completely random measures
• Modeling with normalized random measure mixture models
• Fast Search and Estimation of Bayesian Nonparametric Mixture Models Using a Classification Annealing EM Algorithm
• An adaptive truncation method for inference in Bayesian nonparametric models
• Truncated two-parameter Poisson-Dirichlet approximation for Pitman-Yor process hierarchical models
• Title not available (Why is that?)
• On quasi-infinitely divisible random measures
• MCMC for normalized random measure mixture models

Uses Software

• R
• DPpackage

This page was built for publication: Posterior sampling from \(\epsilon\)-approximation of normalized completely random measure mixtures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q502819)