Predictive construction of priors in Bayesian nonparametrics
From MaRDI portal
Abstract: Predictive constructions are a powerful way of characterizing the probability law of stochastic processes with certain forms of invariance, such as exchangeability or Markov exchangeability. When de Finetti-like representation theorems are available, the predictive characterization implicitly defines the prior distribution, starting from assumptions on the observables; moreover, it often helps designing efficient computational strategies. In this paper we give necessary and sufficient conditions on the sequence of predictive distributions such that they characterize a Markov exchangeable probability law for a discrete valued process X. Under recurrence, Markov exchangeable processes are mixtures of Markov chains. Thus, our results help checking when a predictive scheme characterizes a prior for Bayesian inference on the unknown transition matrix of a Markov chain. Our predictive conditions are in some sense minimal sufficient conditions for Markov exchangeability; we also provide predictive conditions for recurrence. We illustrate their application in relevant examples from the literature and in novel constructions.
Recommendations
Cites work
- scientific article; zbMATH DE number 3738693 (Why is no real title available?)
- scientific article; zbMATH DE number 3518091 (Why is no real title available?)
- scientific article; zbMATH DE number 219860 (Why is no real title available?)
- scientific article; zbMATH DE number 774881 (Why is no real title available?)
- scientific article; zbMATH DE number 3201991 (Why is no real title available?)
- scientific article; zbMATH DE number 2214054 (Why is no real title available?)
- A Bayesian analysis of some nonparametric problems
- A Bayesian nonparametric estimator of a multivariate survival function
- A Glivenko-Cantelli theorem for exchangeable random variables
- A bayesian predictive approach to sequential search for an optimal dose: Parametric and nonparametric models
- A bivariate Dirichlet process.
- A nonparametric urn-based approach to interacting failing systems with an application to credit risk modeling
- A note on extendibility and predictivistic inference in finite populations
- A note on the Dirichlet process prior in Bayesian nonparametric inference with partial exchangeability
- A predictivistic interpretation of the multivariate \(t\)-distribution
- A sticky HDP-HMM with application to speaker diarization
- Bayesian Clustering and Product Partition Models
- Bayesian Density Estimation and Inference Using Mixtures
- Bayesian Nonparametrics
- Bayesian analysis for reversible Markov chains
- Bayesian analysis of variable-order, reversible Markov chains
- Bayesian nonparametrics
- Beta-stacy processes and a generalization of the Pólya-Urn scheme
- Characterizing Markov exchangeable sequences
- Completely random measures
- Conjugate priors for exponential families
- De Finetti's contribution to probability and statistics
- De Finetti's theorem for Markov chains
- Distributional results for means of normalized random measures with independent increments
- Estimating normal means with a conjugate style dirichlet process prior
- Exchangeability, predictive distributions and parametric models
- Exchangeable and partially exchangeable random partitions
- Extendibility of spherical matrix distributions
- Feller operators and mixture priors in Bayesian nonparametrics
- Ferguson distributions via Polya urn schemes
- Gibbs Sampling Methods for Stick-Breaking Priors
- Hierarchical Dirichlet Processes
- Hierarchical Mixture Modeling With Normalized Inverse-Gaussian Priors
- Hierarchical reinforced urn processes
- Hybrid Dirichlet mixture models for functional data
- Invariants Under Mixing Which Generalize de Finetti's Theorem: Continuous Time Parameter
- Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems
- Nonparametric Bayes estimators based on beta processes in models for life history data
- Nonparametric Bayesian data analysis
- On mixtures of distributions of Markov chains.
- On random sequences with spherical symmetry
- On some characterizations of the \(t\)-distribution
- On the support of MacEachern's dependent Dirichlet processes and extensions
- Predictivistic characterizations of multivariate Student-\(t\) models
- Probabilistic Symmetries and Invariance Principles
- Pólya-like urns and the Ewens' sampling formula
- Representations for partially exchangeable arrays of random variables
- Size-biased sampling of Poisson point processes and excursions
- Spherical symmetry: An elementary justification
- Stick-breaking autoregressive processes
- Tailfree and neutral random probabilities and their posterior distributions
- The Representation of Partition Structures
- The sampling theory of selectively neutral alleles
- The two-parameter Poisson-Dirichlet distribution derived from a stable subordinator
- Urn schemes and reinforced random walks.
- W. E. Johnson's sufficientness postulate
Cited in
(8)- The semi-Markov beta-Stacy process: a Bayesian non-parametric prior for semi-Markov processes.
- A de Finetti-type representation of joint hierarchically exchange
- Asymptotics of certain conditionally identically distributed sequences
- Infinite-color randomly reinforced urns with dominant colors
- On a class of prior distributions that accounts for uncertainty in the data
- Bayesian Predictive Inference Without a Prior
- A class of models for Bayesian predictive inference
- Nonsubjective priors via predictive relative entropy regret
This page was built for publication: Predictive construction of priors in Bayesian nonparametrics
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q447985)
