Clustering dynamics in a class of normalised generalised gamma dependent priors

DOI10.1007/S10463-016-0583-8MaRDI QIDQ1695756zbMATH OpenOpenAlexFDO

Authors Peng Zhang

Publication date 8 February 2018

Published in Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1608.00733

zbMATH Keywords

alpha diversityBayesian nonparametricsclusteringdependent processdiffusion processgeneralyzed Pólya urnMoran modelscaling limit

Mathematics Subject Classification ID

Generalized stochastic processes (60G20) Random measures (60G57) Diffusion processes (60J60) Nonparametric estimation (62G05) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Abstract: Normalised generalised gamma processes are random probability measures that induce nonparametric prior distributions widely used in Bayesian statistics, particularly for mixture modelling. We construct a class of dependent normalised generalised gamma priors induced by a stationary population model of Moran type, which exploits a generalised P'olya urn scheme associated with the prior. We study the asymptotic scaling for the dynamics of the number of clusters in the sample, which in turn provides a dynamic measure of diversity in the underlying population. The limit is formalised to be a positive nonstationary diffusion process which falls outside well known families, with unbounded drift and an entrance boundary at the origin. We also introduce a new class of stationary positive diffusions, whose invariant measures are explicit and have power law tails, which approximate weakly the scaling limit.

Review text

Normalized generalized gamma processes induce nonparametric prior distributions widely used for mixture modelling in Bayesian statistics. In the paper under review, a class of dependent normalized generalized gamma priors induced by a stationary population model of Moran type is constructed. The asymptotic scaling for the dynamics of the number of clusters in the sample is studied, which in turn provides a dynamic measure of diversity in the underlying population. The limit is a positive non-stationary diffusion process, with unbounded drift and an entrance boundary at the origin. A new class of stationary positive diffusions is introduced, whose invariant measures are explicit and have power law tails, which approximate weakly the scaling limit.

Recommendations

• A note on Bayesian nonparametric priors derived from exponentially tilted Poisson-Kingman models
• Bayesian inference with dependent normalized completely random measures
• Normalized random measures driven by increasing additive processes
• Asymptotics for a Bayesian nonparametric estimator of species variety
• Clustering and mixtures of processes

Cites work

• A Bayesian analysis of some nonparametric problems
• Alpha-diversity processes and normalized inverse-Gaussian diffusions
• Bayesian inference with dependent normalized completely random measures
• Bayesian nonparametric inference for discovery probabilities: credible intervals and large sample asymptotics
• Combinatorial Methods in Discrete Distributions
• Comparing Distributions by using Dependent Normalized Random-Measure Mixtures
• Compound random measures and their use in Bayesian non-parametrics
• Conjugacy properties of time-evolving Dirichlet and gamma random measures
• Contributions to the theory of Dirichlet processes
• Controlling the Reinforcement in Bayesian Non-Parametric Mixture Models
• Countable representation for infinite dimensional diffusions derived from the two-parameter Poisson-Dirichlet process
• Discreteness of Ferguson selections
• Dynamic density estimation with diffusive Dirichlet mixtures
• Exchangeable and partially exchangeable random partitions
• Generalized Gamma measures and shot-noise Cox processes
• Geometric stick-breaking processes for continuous-time Bayesian nonparametric modeling
• Hierarchical Mixture Modeling With Normalized Inverse-Gaussian Priors
• scientific article; zbMATH DE number 3951715 (Why is no real title available?)
• scientific article; zbMATH DE number 3736679 (Why is no real title available?)
• scientific article; zbMATH DE number 1324455 (Why is no real title available?)
• scientific article; zbMATH DE number 472921 (Why is no real title available?)
• scientific article; zbMATH DE number 5586084 (Why is no real title available?)
• scientific article; zbMATH DE number 2214054 (Why is no real title available?)
• Investigating nonparametric priors with Gibbs structure
• On a class of Bayesian nonparametric estimates: I. Density estimates
• Some mathematical models from population genetics. École d'Été de Probabilités de Saint-Flour XXXIX-2009
• Species dynamics in the two-parameter Poisson-Dirichlet diffusion model
• The Poisson-Dirichlet distribution and related topics. Models and asymptotic behaviors
• The two-parameter Poisson-Dirichlet distribution derived from a stable subordinator
• Two-parameter family of infinite-dimensional diffusions on the Kingman simplex
• Wright-Fisher construction of the two-parameter Poisson-Dirichlet diffusion

Cited in

(3)

• A note on Bayesian nonparametric priors derived from exponentially tilted Poisson-Kingman models
• Markov stick-breaking processes
• On a Gibbs characterization of normalized generalized gamma processes

This page was built for publication: Clustering dynamics in a class of normalised generalised gamma dependent priors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1695756)