Dynamic density estimation with diffusive Dirichlet mixtures

DOI10.3150/14-BEJ681zbMATH Open1388.62099arXiv1410.2477OpenAlexW3103587541MaRDI QIDQ265276FDOQ265276

Authors: Ramsés H. Mena, Matteo Ruggiero 

Publication date: 1 April 2016

Published in: Bernoulli (Search for Journal in Brave)

Abstract: We introduce a new class of nonparametric prior distributions on the space of continuously varying densities, induced by Dirichlet process mixtures which diffuse in time. These select time-indexed random functions without jumps, whose sections are continuous or discrete distributions depending on the choice of kernel. The construction exploits the widely used stick-breaking representation of the Dirichlet process and induces the time dependence by replacing the stick-breaking components with one-dimensional Wright-Fisher diffusions. These features combine appealing properties of the model, inherited from the Wright-Fisher diffusions and the Dirichlet mixture structure, with great flexibility and tractability for posterior computation. The construction can be easily extended to multi-parameter GEM marginal states, which include, for example, the Pitman--Yor process. A full inferential strategy is detailed and illustrated on simulated and real data.

Full work available at URL: https://arxiv.org/abs/1410.2477

Recommendations

• Bayesian dynamic density estimation
• Generalized Pólya urn for time-varying Pitman-Yor processes
• Conjugacy properties of time-evolving Dirichlet and gamma random measures
• Dependent mixtures of Dirichlet processes
• Bayesian Density Estimation and Inference Using Mixtures

zbMATH Keywords

density estimationnonparametric regressionhidden Markov modelDirichlet processPitman-Yor processWright-Fisher diffusion

Mathematics Subject Classification ID

Bayesian inference (62F15) Density estimation (62G07) Diffusion processes (60J60) Random measures (60G57)

Cites Work

• Inference from iterative simulation using multiple sequences
• On the support of MacEachern's dependent Dirichlet processes and extensions
• Nonparametric Bayesian models through probit stick-breaking processes
• Sampling the Dirichlet Mixture Model with Slices
• Generalized Spatial Dirichlet Process Models
• Kernel stick-breaking processes
• An ANOVA Model for Dependent Random Measures
• Title not available (Why is that?)
• Bayesian Nonparametric Spatial Modeling With Dirichlet Process Mixing
• Order-Based Dependent Dirichlet Processes
• A Bayesian analysis of some nonparametric problems
• Bayesian dynamic modeling of latent trait distributions
• Hierarchical Mixture Modeling With Normalized Inverse-Gaussian Priors
• The two-parameter Poisson-Dirichlet distribution derived from a stable subordinator
• Exchangeable and partially exchangeable random partitions
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Slice sampling mixture models
• Bayesian Nonparametrics
• Retrospective Markov chain Monte Carlo methods for Dirichlet process hierarchical models
• Gibbs Sampling Methods for Stick-Breaking Priors
• On a class of Bayesian nonparametric estimates: I. Density estimates
• A nonparametric dependent process for Bayesian regression
• Diffusion-type models with given marginal distribution and autocorrelation function
• Bayesian nonparametrics
• Alpha-diversity processes and normalized inverse-Gaussian diffusions
• On a Gibbs sampler based random process in Bayesian nonparametrics
• Bayesian dynamic density estimation
• Hybrid Dirichlet mixture models for functional data
• The multivariate beta process and an extension of the Pólya tree model
• The Matrix Stick-Breaking Process
• The infinitely-many-neutral-alleles diffusion model
• Gibbs Sampling for Bayesian Non-Conjugate and Hierarchical Models by Using Auxiliary Variables
• Bayesian Inference for Linear Dynamic Models With Dirichlet Process Mixtures
• Fleming–Viot Processes in Population Genetics
• Bayesian Density Regression
• A Class of Infinite-Dimensional Diffusion Processes with Connection to Population Genetics
• Geometric stick-breaking processes for continuous-time Bayesian nonparametric modeling
• The transition function of a Fleming-Viot process
• Stick-breaking autoregressive processes
• Two-parameter family of infinite-dimensional diffusions on the Kingman simplex
• Countable representation for infinite dimensional diffusions derived from the two-parameter Poisson-Dirichlet process

Cited In (14)

• The dependent Dirichlet process and related models
• Bayesian temporal density estimation with autoregressive species sampling models
• Innovation, growth and aggregate volatility from a Bayesian nonparametric perspective
• Latent nested nonparametric priors (with discussion)
• Generalized Pólya urn for time-varying Pitman-Yor processes
• Computational challenges and temporal dependence in Bayesian nonparametric models
• Fast dynamic nonparametric distribution tracking in electron microscopic data
• Predictive inference with Fleming-Viot-driven dependent Dirichlet processes
• Bayesian prediction with multiple-samples information
• Direct deconvolution density estimation of a mixture distribution motivated by mutation effects distribution
• Dm-KDE: dynamical kernel density estimation by sequences of KDE estimators with fixed number of components over data streams
• Strong convergence and stationary distribution of an explicit scheme for the Wright-Fisher model
• Conjugacy properties of time-evolving Dirichlet and gamma random measures
• Clustering dynamics in a class of normalised generalised gamma dependent priors

This page was built for publication: Dynamic density estimation with diffusive Dirichlet mixtures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q265276)