BAYESIAN GENERALIZED PRODUCT PARTITION MODEL
DukeSpace (Duke University)Published 1 July 2010Open access
Ju‐Hyun Park, David B. Dunson
Citations58
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A generalized product partition model (GPPM) in which the parti- tion process is predictor-dependent is derived, which generalizes DP clustering to relax the exchangeability assumption through the incorporation of predictors, resulting in a generalized Polyaurn scheme.
Abstract
Starting with a carefully formulated Dirichlet process (DP) mixture model, we derive a generalized product partition model (GPPM) in which the partition process is predictor-dependent. The GPPM generalizes DP clustering to relax the exchangeability assumption through the incorporation of predictors, resulting in a generalized Pólya urn scheme. In addition, the GPPM can be used for formulating flexible semiparametric Bayes models for conditional distribution estimation, bypassing the need for expensive computation of large numbers of unknowns characterizing priors for dependent collections of random probability measures. Properties are discussed, a variety of special cases are considered, and an effi-cient Gibbs sampling algorithm is developed for posterior computation. The methods are illustrated using simulation examples and an epidemiologic application.
Keywords
Computer ScienceMathematics
BiometrikaMonte Carlo sampling methods using Markov chains and their applications
15,200 Citations1970W. Keith Hastings
The Annals of StatisticsA Bayesian Analysis of Some Nonparametric Problems
4,812 Citations1973Thomas S. Ferguson
Journal of the American Statistical AssociationModel-Based Clustering, Discriminant Analysis, and Density Estimation
4,282 Citations2002Chris Fraley, Adrian E. Raftery
This work reviews a general methodology for model-based clustering that provides a principled statistical approach to important practical questions that arise in cluster analysis, such as how many clusters are there, which clustering method should be used, and how should outliers be handled.
Neural ComputationHierarchical Mixtures of Experts and the EM Algorithm
2,639 Citations1994Michael I. Jordan, Robert A. Jacobs
Journal of the American Statistical AssociationBayesian Density Estimation and Inference Using Mixtures
2,241 Citations1995Michael Escobar, Mike West
These models provide natural settings for density estimation and are exemplified by special cases where data are modeled as a sample from mixtures of normal distributions, allowing for direct inference on a variety of practical issues.
Journal of Computational and Graphical StatisticsMarkov Chain Sampling Methods for Dirichlet Process Mixture Models
2,212 Citations2000Radford M. Neal
The Annals of StatisticsFerguson Distributions Via Polya Urn Schemes
1,346 Citations1973David Blackwell, James B. MacQueen
Journal of the Royal Statistical Society Series B (Statistical Methodology)Dealing With Label Switching in Mixture Models
1,088 Citations2000Matthew Stephens
It is demonstrated that this fails in general to solve the ‘label switching’ problem, and an alternative class of approaches, relabelling algorithms, which arise from attempting to minimize the posterior expected loss under a class of loss functions are described.
The Annals of StatisticsPrior Distributions on Spaces of Probability Measures
774 Citations1974Thomas S. Ferguson
The Annals of StatisticsOn a Class of Bayesian Nonparametric Estimates: I. Density Estimates
738 Citations1984Albert Y. Lo
Journal of the American Statistical AssociationComputational and Inferential Difficulties with Mixture Posterior Distributions
568 Citations2000Gilles Celeux, Merrilee Hurn +1 more
This article proposes alternatives for Bayesian inference for permutation invariant posteriors, including a clustering device and alternative appropriate loss functions and shows that exploration of these modes can be imposed using tempered transitions.
Journal of the American Statistical AssociationEstimating Normal Means with a Dirichlet Process Prior
539 Citations1994Michael Escobar
Journal of Computational and Graphical StatisticsEstimating Mixture of Dirichlet Process Models
539 Citations1998Steven N. MacEachern, Peter Müller
A conceptual framework for computational strategies is proposed that provides a perspective on current methods, facilitates comparisons between them, and leads to several new methods that expand the scope of MDP models to nonconjugate situations.
The LancetAssociation between maternal serum concentration of the DDT metabolite DDE and preterm and small-for-gestational-age babies at birth
446 Citations2001Matthew P. Longnecker, Mark A. Klebanoff +2 more
The findings strongly suggest that DDT use increases preterm births, which is a major contributor to infant mortality, and should be included in any assessment of the costs and benefits of vector control with DDT.
Communications in Statistics - Simulation and ComputationEstimating normal means with a conjugate style dirichlet process prior
421 Citations1994Steven N. MacEachern
Lecture notes-monograph seriesSome developments of the Blackwell-MacQueen urn scheme
418 Citations1996Jim Pitman
BiometrikaRetrospective Markov chain Monte Carlo methods for Dirichlet process hierarchical models
362 Citations2008Omiros Papaspiliopoulos, Gareth O. Roberts
This paper designs two novel Markov chain Monte Carlo algorithms which sample from the exact posterior distribution of quantities of interest and shows how the algorithms can obtain samples from functionals of the Dirichlet process.
Journal of the American Statistical AssociationBayesian Nonparametric Spatial Modeling With Dirichlet Process Mixing
361 Citations2005Alan E. Gelfand, Athanasios Kottas +1 more
This work develops a spatial Dirichlet process model for spatial data and discusses its properties, and introduces mixing by convolving this process with a pure error process to produce a random spatial process that is neither Gaussian nor stationary.
Journal of the American Statistical AssociationOrder-Based Dependent Dirichlet Processes
359 Citations2006Jim E. Griffin, Mark F. J. Steel
This article allows the nonparametric distribution to depend on covariates through ordering the random variables building the weights in the stick-breaking representation and derives the correlation between distributions at different covariate values.
Journal of the American Statistical AssociationAn ANOVA Model for Dependent Random Measures
348 Citations2004Maria De Iorio, Peter Müller +2 more
A model that describes dependence across random distributions in an analysis of variance (ANOVA)-type fashion is proposed that can be rewritten as a DP mixture of ANOVA models, which inherits all computational advantages of standard DP mixture models.
BiometrikaA semiparametric Bayesian model for randomised block designs
309 Citations1996Christopher Bush
Cambridge University Press eBooksModel-Based Clustering for Expression Data via a Dirichlet Process Mixture Model
298 Citations2006David B. Dahl
This chapter introduces a method to get a point estimate of the true clustering based on least-squares distances from the posterior probability that two genes are clustered, and compared to other clustering methods in a simulation study.
BiometrikaKernel stick-breaking processes
284 Citations2008David B. Dunson, Ji‐Hye Park
A retrospective Markov chain Monte Carlo algorithm is developed for posterior computation, and the methods are illustrated using a simulated example and an epidemiological application.
The Annals of StatisticsProduct Partition Models for Change Point Problems
283 Citations1992Daniel Barry, J. A. Hartigan
Journal of the Royal Statistical Society Series B (Statistical Methodology)Bayesian Density Regression
263 Citations2007David B. Dunson, Natesh S. Pillai +1 more
The paper considers Bayesian methods for density regression, allowing a random probability distribution to change flexibly with multiple predictors, and proposes a kernel‐based weighting scheme that incorporates weights that are dependent on the distance between subjects’ predictor values.
Journal of Computational and Graphical StatisticsBayesian Model-Based Clustering Procedures
222 Citations2007John W. Lau, Peter J. Green
This article establishes a general formulation for Bayesian model-based clustering, in which subset labels are exchangeable, and items are also Exchangeable, possibly up to covariate effects, and a new heuristic item-swapping algorithm is introduced.
Canadian Journal of StatisticsSequential importance sampling for nonparametric Bayes models: The next generation
200 Citations1999Steven N. MacEachern, Merlise A. Clyde +1 more
It is shown that integration and appending a second stage to the Gibbs sampler wherein the cluster locations are moved dramatically improves results and is shown to provide more uniform importance sampling weights and lead to additional Rao‐Blackwellization of estimators.
Journal of the Royal Statistical Society Series B (Statistical Methodology)Bayesian Clustering and Product Partition Models
196 Citations2003Fernando A. Quintana, Pilar L. Iglesias
A decision theoretic formulation of product partition models (PPMs) is presented that allows a formal treatment of different decision problems such as estimation or hypothesis testing and clustering methods simultaneously, and an algorithm is proposed that yields Bayes estimates of the quantities of interest and the groups of experimental units.
Generalized weighted Chinese restaurant processes for species sampling mixture models
161 Citations2003Hemant Ishwaran, Lancelot F. James
The Annals of StatisticsHierarchical mixtures-of-experts for exponential family regression models: approximation and maximum likelihood estimation
144 Citations1999Wenxin Jiang, Martin A. Tanner
The MIT Press eBooksAccelerated Variational Dirichlet Process Mixtures
118 Citations2007Kenichi Kurihara, Max Welling +1 more
AddictionAlcohol dependence: a public health perspective
109 Citations2002Raúl Caetano, Carol B. Cunradi
Alcohol dependence is present at relatively high levels in the community and a public health approach, emphasizing prevention and group level interventions to lower prevalence levels, is warranted.
Journal of the American Statistical AssociationExtending the Akaike Information Criterion to Mixture Regression Models
103 Citations2007P. A. Naik, Peide Shi +1 more
Journal of Statistical Planning and InferenceA predictive view of Bayesian clustering
66 Citations2004Fernando A. Quintana
This work considers probability models for partitions of a set of n elements using a predictive approach, i.e., models that are specified in terms of the conditional probability of either joining an already existing cluster or forming a new one.
BiometrikaBayesian nonparametric inference on stochastic ordering
46 Citations2008David B. Dunson, Shyamal D. Peddada
Canadian Journal of StatisticsModeling nonlinear time series with local mixtures of generalized linear models
45 Citations2005Alexandre Xavier Ywata de Carvalho, Martin A. Tanner
Journal of the Science of Food and AgricultureDecision Theoretic Aspects of Dependent Nonparametric Processes
41 Citations2000Steven N. MacEachern
The technique can be used to assay protein amylase inhibitors from wheat at concentrations as low as 1.7 nM and may also beUsed to detect amyl enzyme activity which can continue in spite of the presence of inhibitors.
Journal of Computational and Graphical StatisticsComputational Aspects of Nonparametric Bayesian Analysis with Applications to the Modeling of Multiple Binary Sequences
32 Citations2000Fernando A. Quintana, Michael A. Newton
The findings suggest that the sequential imputations method is most useful for relatively small problems, and that the predictive recursion can be an efficient preliminary tool for more reliable, but computationally intensive, Gibbs sampling implementations.
Bayesian Inference for Dynamic Models with Dirichlet Process Mixtures
28 Citations2006François Caron, Manuel Davy +3 more
A flexible Bayesian nonparametric noise model based on mixture of Dirichlet processes is introduced and efficient Markov chain Monte Carlo and sequential Monte Carlo methods are developed to perform optimal estimation in linear Gaussian state-space models.
Neural ComputationOn Consistency of Bayesian Inference with Mixtures of Logistic Regression
20 Citations2005Yang Ge, Wenxin Jiang
A theoretical study of the consistency properties of Bayesian inference using mixtures of logistic regression models, which shows these desirable consistency properties with a nonstochastic k growing slowly with the sample size n of the observed data, or with a random k that takes large values with nonzero but small probabilities.
Journal of Computational and Graphical StatisticsBayesian Prediction via Partitioning
18 Citations2005Chris Holmes, D. G. T. Denison +2 more
A new Bayesian approach to prediction on continuous covariates is proposed, which constructs arbitrarily complex regression and classification surfaces by splitting the covariate space into an unknown number of disjoint regions.
arXiv (Cornell University)Retrospective Markov chain Monte Carlo methods for Dirichlet process hierarchical model
9 Citations2007Omiros Papaspiliopoulos, Gareth O. Roberts