Confronting prior convictions: On issues of prior sensitivity and likelihood robustness in bayesian analysis

Hedibert F. Lopes, Justin L. Tobias

Research output: Contribution to journalReview articlepeer-review

11 Scopus citations

Abstract

In this review we explore issues of the sensitivity of Bayes estimates to the prior and form of the likelihood. With respect to the prior, we argue that non-Bayesian analyses also incorporate prior information, illustrate that the Bayes posterior mean and the frequentist maximum likelihood estimator are often asymptotically equivalent, review a simple computational strategy for analyzing sensitivity to the prior in practice, and finally document the potentially important role of the prior in Bayesian model comparison. With respect to issues of likelihood robustness, we review a variety of computational strategies for significantly expanding the maintained sampling model, including the use of finite Gaussian mixture models and models based on Dirichlet process priors.

Original languageEnglish (US)
Pages (from-to)107-131
Number of pages25
JournalAnnual Review of Economics
Volume3
DOIs
StatePublished - 2011
Externally publishedYes

Keywords

  • Bayesian methods
  • Dirichlet process mixture
  • Factor models
  • Gibbs sampler
  • Marginal likelihood
  • Markov chain Monte Carlo
  • Scale mixture of normals
  • Sequential Monte Carlo

ASJC Scopus subject areas

  • Economics and Econometrics

Fingerprint

Dive into the research topics of 'Confronting prior convictions: On issues of prior sensitivity and likelihood robustness in bayesian analysis'. Together they form a unique fingerprint.

Cite this