Bayesian variance estimation in the Gaussian sequence model with partial information on the means

DOI10.1214/19-EJS1671zbMATH Open1454.62097arXiv1904.04525MaRDI QIDQ2286367FDOQ2286367

Johannes Schmidt-Hieber, Gianluca Finocchio

Publication date: 22 January 2020

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: Consider the Gaussian sequence model under the additional assumption that a fixed fraction of the means is known. We study the problem of variance estimation from a frequentist Bayesian perspective. The maximum likelihood estimator (MLE) for $sigm{a}^2$ is biased and inconsistent. This raises the question whether the posterior is able to correct the MLE in this case. By developing a new proving strategy that uses refined properties of the posterior distribution, we find that the marginal posterior is inconsistent for any i.i.d. prior on the mean parameters. In particular, no assumption on the decay of the prior needs to be imposed. Surprisingly, we also find that consistency can be retained for a hierarchical prior based on Gaussian mixtures. In this case we also establish a limiting shape result and determine the limit distribution. In contrast to the classical Bernstein-von Mises theorem, the limit is non-Gaussian. We show that the Bayesian analysis leads to new statistical estimators outperforming the correctly calibrated MLE in a numerical simulation study.

Full work available at URL: https://arxiv.org/abs/1904.04525

Recommendations

• Adaptive Bayesian inference in the Gaussian sequence model using exponential-variance priors
• Variance estimation with diffuse prior information
• Adaptive Bayesian estimation in Gaussian sequence space models
• Consistency of Bayesian estimates for the sum of squared normal means with a normal prior
• Small-sample inference about variance and its transformations

zbMATH Keywords

maximum likelihoodsemiparametric inferencefrequentist BayesGaussian sequence modelBernstein-von Mises theorems

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Bayesian inference (62F15)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Consistent Estimates Based on Partially Consistent Observations
• Consistency of the Maximum Likelihood Estimator in the Presence of Infinitely Many Incidental Parameters
• Fundamentals of Nonparametric Bayesian Inference
• Asymptotic methods in statistical decision theory
• Likelihood Estimation and Inference in a Class of Nonregular Econometric Models
• The multivariate skew-normal distribution
• A semiparametric Bernstein-von Mises theorem for Gaussian process priors
• A Bernstein-von Mises theorem for smooth functionals in semiparametric models
• The semiparametric Bernstein-von Mises theorem
• Lower bounds for posterior rates with Gaussian process priors
• Needles and straw in a haystack: posterior concentration for possibly sparse sequences
• Conditions for posterior contraction in the sparse normal means problem
• Symmetric Measures on Cartesian Products
• Asymptotic Normality of Semiparametric and Nonparametric Posterior Distributions
• Bayesian linear regression with sparse priors
• On adaptive posterior concentration rates
• Semiparametric Bernstein-von Mises for the error standard deviation
• Semiparametric Bernstein-von Mises theorem and bias, illustrated with Gaussian process priors
• Asymptotic behaviour of Bayes estimates and posterior distributions in multiparameter nonregular cases
• Title not available (Why is that?)
• Empirical Bayes analysis of spike and slab posterior distributions
• Testing in high-dimensional spiked models
• Nonparametric Bayesian analysis of the compound Poisson prior for support boundary recovery

This page was built for publication: Bayesian variance estimation in the Gaussian sequence model with partial information on the means

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2286367)