Evaluating default priors with a generalization of Eaton's Markov chain

DOI10.1214/13-AIHP552zbMATH Open1298.62016arXiv0912.4566MaRDI QIDQ405509FDOQ405509

Authors: Brian P. Shea, Galin L. Jones 

Publication date: 5 September 2014

Published in: Annales de l'Institut Henri Poincaré. Probabilités et Statistiques (Search for Journal in Brave)

Abstract: We consider evaluating improper priors in a formal Bayes setting according to the consequences of their use. Let $Phi$ be a class of functions on the parameter space and consider estimating elements of $Phi$ under quadratic loss. If the formal Bayes estimator of every function in $Phi$ is admissible, then the prior is strongly admissible with respect to $Phi$. Eaton's method for establishing strong admissibility is based on studying the stability properties of a particular Markov chain associated with the inferential setting. In previous work, this was handled differently depending upon whether $phiinPhi$ was bounded or unbounded. We introduce and study a new Markov chain which allows us to unify and generalize existing approaches while simultaneously broadening the scope of their potential applicability. To illustrate the method, we establish strong admissibility conditions when the model is a $p$-dimensional multivariate normal distribution with unknown mean vector $heta$ and the prior is of the form $u(|heta{|}^2)dheta$.

Full work available at URL: https://arxiv.org/abs/0912.4566

Recommendations

• A decision-theoretical view of default priors
• A new approach to default priors and robust bayes methodology
• scientific article; zbMATH DE number 2127975
• Default prior distributions from quasi- and quasi-profile likelihoods
• De Finetti priors using Markov chain Monte Carlo computations
• Evaluation of formal posterior distributions via Markov chain arguments
• A Bayesian Model for Markov Chains via Jeffrey's Prior
• On the use of Markovian stick-breaking priors
• On distributions with fixed marginals maximizing the joint or the prior default probability, estimation, and related results

zbMATH Keywords

recurrenceadmissibilityDirichlet formformal Bayes ruleimproper prior distributionsymmetric Markov chain

Mathematics Subject Classification ID

Discrete-time Markov processes on general state spaces (60J05) Admissibility in statistical decision theory (62C15)

Cites Work

• The Selection of Prior Distributions by Formal Rules
• Markov chains and stochastic stability
• Estimation with quadratic loss.
• Title not available (Why is that?)
• Title not available (Why is that?)
• Choice of hierarchical priors: Admissibility in estimation of normal means
• A statistical diptych: Admissible inferences -- recurrence of symmetric Markov chains
• Admissibility in quadratically regular problems and recurrence of symmetric Markov chains: Why the connection?
• Eaton's Markov chain, its conjugate partner and \(\mathcal P\)-admissibility
• Conditions for recurrence and transience of a Markov chain on \(\mathbb Z^ +\) and estimation of a geometric success probability.
• On perturbations of strongly admissible prior distributions
• Posterior propriety and admissibiity of hyperpriors in normal hierarchical models
• When is Eaton's Markov chain irreducible?
• Stein estimation for spherically symmetric distributions: recent developments
• The Admissibility of Pitman's Estimator of a Single Location Parameter
• Title not available (Why is that?)
• Evaluating improper priors and the recurrence of symmetric Markov chains: an overview
• Markov chain conditions for admissibility in estimation problems with quadratic loss
• Admissible Estimators, Recurrent Diffusions, and Insoluble Boundary Value Problems
• Evaluation of formal posterior distributions via Markov chain arguments

Cited In (6)

• Stability of the tail Markov chain and the evaluation of improper priors for an exponential rate parameter
• Evaluation of formal posterior distributions via Markov chain arguments
• Evaluating improper priors and the recurrence of symmetric Markov chains: an overview
• On perturbations of strongly admissible prior distributions
• When is Eaton's Markov chain irreducible?
• Recurrence and transience of a Markov chain on \(\mathbb{Z}^+\) and evaluation of prior distributions for a Poisson mean

This page was built for publication: Evaluating default priors with a generalization of Eaton's Markov chain

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q405509)