On posterior concentration in misspecified models

DOI10.1214/15-BA941MaRDI QIDQ273635zbMATHFDO

Authors R. V. Ramamoorthi, Karthik Sriram, Ryan Martin

Publication date 22 April 2016

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1312.4620

zbMATH Keywords

BayesianconsistencyKullback-Leiblermisspecified

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Nonparametric regression and quantile regression (62G08) Asymptotic distribution theory in statistics (62E20) Bayesian problems; characterization of Bayes procedures (62C10) Strong limit theorems (60F15)

Abstract: We investigate the asymptotic behavior of Bayesian posterior distributions under independent and identically distributed ($i.i.d.$) misspecified models. More specifically, we study the concentration of the posterior distribution on neighborhoods of $f^{star}$, the density that is closest in the Kullback--Leibler sense to the true model $f_0$. We note, through examples, the need for assumptions beyond the usual Kullback--Leibler support assumption. We then investigate consistency with respect to a general metric under three assumptions, each based on a notion of divergence measure, and then apply these to a weighted $L_1$-metric in convex models and non-convex models. Although a few results on this topic are available, we believe that these are somewhat inaccessible due, in part, to the technicalities and the subtle differences compared to the more familiar well-specified model case. One of our goals is to make some of the available results, especially that of , more accessible. Unlike their paper, our approach does not require construction of test sequences. We also discuss a preliminary extension of the $i.i.d.$ results to the independent but not identically distributed ($i.n.i.d.$) case.

Recommendations

• scientific article; zbMATH DE number 6142324
• Misspecification in infinite-dimensional Bayesian statistics
• Asymptotic properties of posterior distributions derived from misspecified models
• Posterior concentration for a misspecified Bayesian regression model with functional covariates
• On Rates of Convergence for Posterior Distributions Under Misspecification

Cites work

• scientific article; zbMATH DE number 6142324 (Why is no real title available?)
• Asymptotic behavior of Bayes estimates under possibly incorrect models
• Bayesian nonparametrics
• Bayesian quantile regression
• Consistency of Bayes estimators without the assumption that the model is correct
• Dynamics of Bayesian updating with dependent data and misspecified models
• Limiting Behavior of Posterior Distributions when the Model is Incorrect
• Misspecification in infinite-dimensional Bayesian statistics
• On general minimax theorems
• Posterior consistency of Bayesian quantile regression based on the misspecified asymmetric Laplace density

Cited in

(21)

• Posterior concentration for a misspecified Bayesian regression model with functional covariates
• General Robust Bayes Pseudo-Posteriors: Exponential Convergence Results with Applications
• On Rates of Convergence for Posterior Distributions Under Misspecification
• On Bayesian quantile regression using a pseudo-joint asymmetric Laplace likelihood
• Analyzing posteriors by the information inequality
• Model Misspecification in Approximate Bayesian Computation: Consequences and Diagnostics
• The Bernstein-von Mises theorem under misspecification
• A note on a posterior approximation in a heteroscedastic model
• Bayesian fractional posteriors
• Pathwise concentration bounds for Bayesian beliefs
• Asymptotic properties of posterior distributions derived from misspecified models
• About the non-asymptotic behaviour of Bayes estimators
• Model misspecification, Bayesian versus credibility estimation, and Gibbs posteriors
• Robust sparse Bayesian infinite factor models
• Misspecification in infinite-dimensional Bayesian statistics
• What About the Posterior Distributions When the Model is Non-dominated?
• A review of Bayesian asymptotics in general insurance applications
• A comparison of learning rate selection methods in generalized Bayesian inference
• Contextuality of misspecification and data-dependent losses
• Gibbs posterior concentration rates under sub-exponential type losses
• Gibbs posterior inference on the minimum clinically important difference

This page was built for publication: On posterior concentration in misspecified models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q273635)