Misspecification in infinite-dimensional Bayesian statistics

DOI10.1214/009053606000000029MaRDI QIDQ2497184zbMATH OpenOpenAlexFDO

Authors B. J. K. Kleijn, Aad van der Vaart

Publication date 3 August 2006

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0607023

zbMATH Keywords

infinite-dimensional modelmisspecificationposterior distributionrate of convergence

Mathematics Subject Classification ID

Foundations and philosophical topics in statistics (62A01) Statistical aspects of information-theoretic topics (62B10) Asymptotic properties of parametric tests (62F05) Bayesian inference (62F15) Density estimation (62G07) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20)

Abstract: We consider the asymptotic behavior of posterior distributions if the model is misspecified. Given a prior distribution and a random sample from a distribution $P_0$, which may not be in the support of the prior, we show that the posterior concentrates its mass near the points in the support of the prior that minimize the Kullback--Leibler divergence with respect to $P_0$. An entropy condition and a prior-mass condition determine the rate of convergence. The method is applied to several examples, with special interest for infinite-dimensional models. These include Gaussian mixtures, nonparametric regression and parametric models.

Recommendations

• scientific article; zbMATH DE number 6142324
• On posterior concentration in misspecified models
• Asymptotic properties of posterior distributions derived from misspecified models
• On Rates of Convergence for Posterior Distributions Under Misspecification
• Convergence rates of posterior distributions.

Cites work

• A Bayesian analysis of some nonparametric problems
• An Introduction to Banach Space Theory
• Approximation dans les espaces m�triques et th�orie de l'estimation
• Asymptotic behavior of Bayes estimates under possibly incorrect models
• Asymptotic methods in statistical decision theory
• Consistency of maximum likelihood estimators for certain nonparametric families, in particular: Mixtures
• Convergence rates of posterior distributions.
• Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities.
• Limiting Behavior of Posterior Distributions when the Model is Incorrect
• Mathematical theory of statistics. Statistical experiments and asymptotic decision theory
• On Bayes procedures
• On inconsistent Bayes estimates of location
• On the consistency of Bayes estimates
• Prior distributions on spaces of probability measures
• Probability inequalities for likelihood ratios and convergence rates of sieve MLEs
• Rates of convergence of posterior distributions.

Cited in

(87)

• A note on Bayesian nonparametric regression function estimation
• Bayesian model averaging for multivariate extremes
• A theoretical note on the prior information criterion
• A review of Bayesian asymptotics in general insurance applications
• Bayesian fractional posteriors
• Asymptotic properties of posterior distributions derived from misspecified models
• The Bernstein-von Mises theorem under misspecification
• Dynamics of Bayesian updating with dependent data and misspecified models
• Asymptotic properties of predictive recursion: robustness and rate of convergence
• On posterior contraction of parameters and interpretability in Bayesian mixture modeling
• Robust Bayes-like estimation: rho-Bayes estimation
• Criteria for posterior consistency and convergence at a rate
• A survey of nonparametric mixing density estimation via the predictive recursion algorithm
• Oracle posterior contraction rates under hierarchical priors
• Robust sparse Bayesian infinite factor models
• An inverse Sanov theorem for exponential families
• Community detection for weighted networks with unknown number of communities
• Consistency of Bayesian inference for multivariate max-stable distributions
• Model-free posterior inference on the area under the receiver operating characteristic curve
• Partially censored posterior for robust and efficient risk evaluation
• Bayesian shrinkage towards sharp minimaxity
• Bayesian fusion estimation via \(t\) shrinkage
• A general framework for Bayes structured linear models
• Posterior concentration for a misspecified Bayesian regression model with functional covariates
• On an algorithm for solving Fredholm integrals of the first kind
• Bayesian regression with nonparametric heteroskedasticity
• Suboptimal behavior of Bayes and MDL in classification under misspecification
• Gibbs posterior inference on the minimum clinically important difference
• Nonparametric Bayesian model selection and averaging
• Quasi-Bayesian analysis of nonparametric instrumental variables models
• Bayesian variable selection for high dimensional generalized linear models: convergence rates of the fitted densities
• Brittleness of Bayesian inference under finite information in a continuous world
• The Bayesian analysis of complex, high-dimensional models: can it be CODA?
• Nearly optimal Bayesian shrinkage for high-dimensional regression
• Gibbs posterior concentration rates under sub-exponential type losses
• Finite sample Bernstein-von Mises theorem for semiparametric problems
• On posterior concentration in misspecified models
• On the brittleness of Bayesian inference
• Posterior contraction rate for non-parametric Bayesian estimation of the dispersion coefficient of a stochastic differential equation
• On Rates of Convergence for Posterior Distributions Under Misspecification
• Bayesian inference with misspecified models
• The semiparametric Bernstein-von Mises theorem
• Critical dimension in the semiparametric Bernstein-von Mises theorem
• Risk minimization for time series binary choice with variable selection
• Fast rates for general unbounded loss functions: from ERM to generalized Bayes
• Quantile graphical models: a Bayesian approach
• Convergence guarantees for Gaussian process means with misspecified likelihoods and smoothness
• General Robust Bayes Pseudo-Posteriors: Exponential Convergence Results with Applications
• Normality of posterior distribution under misspecification and nonsmoothness, and Bayes factor for Davies' problem
• MCMC Computations for Bayesian Mixture Models Using Repulsive Point Processes
• Model misspecification, Bayesian versus credibility estimation, and Gibbs posteriors
• Considering discrepancy when calibrating a mechanistic electrophysiology model
• Fully Bayesian analysis of the relevance vector machine with an extended hierarchical prior structure
• Posterior consistency for the spectral density of non‐Gaussian stationary time series
• Consistency of Bayes estimators without the assumption that the model is correct
• Bayesian Distributionally Robust Optimization
• Bernstein-von Mises theorem and misspecified models: a review
• On accuracy of Gaussian approximation in Bayesian semiparametric problems
• A comparison of learning rate selection methods in generalized Bayesian inference
• Metropolis–Hastings via Classification
• Revisiting consistency of a recursive estimator of mixing distributions
• Pathwise concentration bounds for Bayesian beliefs
• Some models are useful, but how do we know which ones? Towards a unified Bayesian model taxonomy
• Conformal Prediction Credibility Intervals
• Approximate Bayesian computation via classification
• A diffusion process perspective on posterior contraction rates for parameters
• Bayesian learning with multiple priors and nonvanishing ambiguity
• Parametrization, prior independence, and the semiparametric Bernstein-von Mises theorem for the partially linear model
• Directional Gaussian hypergeometric beta distributions and their uses in contaminated binary sampling
• Bayesian mixture models with repulsive and attractive atoms
• Anytime-valid and asymptotically efficient inference driven by predictive recursion
• Statistical foundations of person parameter estimation in the Thurstonian IRT model for forced-choice and pairwise comparison designs
• Generalized median of means principle for Bayesian inference
• Accuracy of Gaussian approximation for high-dimensional posterior distributions
• Large sample justifications for the Bayesian empirical likelihood
• Robust universal inference for misspecified models
• Nonparametric Bayesian inference for ergodic diffusions
• Robust probabilistic inference via a constrained transport metric (with discussion)
• Bayesian Nonparametric Spectral Analysis of Locally Stationary Processes
• Nonparametric Bayesian estimation of a Hölder continuous diffusion coefficient
• Asymptotic equivalence of empirical likelihood and Bayesian MAP
• Local-mass preserving prior distributions for nonparametric Bayesian models
• About the non-asymptotic behaviour of Bayes estimators
• Bernshteĭn-von Mises theorems for nonparametric function analysis via locally constant modelling: a unified approach
• Asymptotic properties of posterior distributions in nonparametric regression with non-Gaussian errors
• Gibbs posterior for variable selection in high-dimensional classification and data mining
• Posterior convergence rates of Dirichlet mixtures at smooth densities

This page was built for publication: Misspecification in infinite-dimensional Bayesian statistics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2497184)