Bayesian robustness for hierarchical ε-contamination models

https://doi.org/10.1016/0378-3758(93)90085-KGet rights and content

Abstract

In the robust Bayesian literature in order to investigate robustness with respect to the functional form of a base prior distribution π0 (in particular with respect to the shape of the prior tails) the ε-contamination model of prior distributions Γ={π: π=(1−ε)π0(θ|λ)+εq,qQ}, has been proposed. Here π0(θ|λ) is the base elicited prior, λ is a vector of fixed hyperparameters, q is a contamination belonging to some suitable class Q and ε reflects the amount of error in π0(θ|λ). For the location-scale family of priors π0(θ|λ=τ-1π0[(θμ)/τ}], where λ=(μ,τ), it turns out that the posterior inference is highly sensitive to changes of λ. Therefore, instead of considering a single base prior π0, it is reasonable to replace π0(θ;|λ) by π0(θ|λ1)=∝π0(θ|λ1,λ2)h(2), where λ1 is the ‘known’ part of λ and λ2 is the ‘uncertain’ part, with the uncertainty reflected by allowing hϵH (H being a class of probability measures). Classes considered here include those with unimodality and quantile constraints. Ranges of posterior probabilities of any set may be computed. Illustrations related to normal hypothesis testing are given.

References (14)

  • J.O. Berger

    Robust Bayesian analysis: Sensitivity to the priors

    J. Statist. Plann. Infer.

    (1990)
  • D.F. Andrews et al.

    Scale mixtures of Normal distributions

    J. Royal Statist. Soc. Series B

    (1974)
  • J.O. Berger

    The robust Bayesian viewpoint (with discussion)

  • J.O. Berger

    Statistical Decision Theory and Bayesian Analysis

    (1985)
  • J.O. Berger et al.

    Robust Bayes and empirical Bayes analysis with ε-contaminated priors

    Ann. Statist.

    (1986)
  • B. de Finetti

    The Bayesian approach to the rejection of outliers

    Proc. 4th Berkeley Symp. on Math. Stat. and Prob.

    (1961)
  • L. DeRobertis et al.

    Bayesian inference using intervals of measures

    Ann. of Statist.

    (1981)
There are more references available in the full text version of this article.

Cited by (12)

  • Accounting for uncertainty in DNA sequencing data

    2015, Trends in Genetics
    Citation Excerpt :

    Methods that can make inferences in the presence of these complicating difficulties are needed. Robust Bayesian [44–49] inference allows for the consideration and analysis of imprecise sample data or ignorance about the appropriate prior assumptions (both of which are the manifestations of epistemic uncertainty and can be modeled using intervals, bounding approaches, and robust Bayesian approaches). In a robust Bayesian analysis, results are considered robust if neither imprecision in the input data nor differences in the prior probability distribution have large effect on the output.

  • Eliciting density ratio classes

    2011, International Journal of Approximate Reasoning
  • Robust Bayesian Analysis

    2005, Handbook of Statistics
  • A simple diagnostic tool for local prior sensitivity

    1997, Statistics and Probability Letters
View all citing articles on Scopus
View full text