Bayesian heavy-tailed models and conflict resolution: A review

  title={Bayesian heavy-tailed models and conflict resolution: A review},
  author={Anthony O'Hagan and Luis Ra{\'u}l Pericchi},
  journal={Brazilian Journal of Probability and Statistics},
  • A. O'Hagan, L. Pericchi
  • Published 1 November 2012
  • Computer Science
  • Brazilian Journal of Probability and Statistics
We review a substantial literature, spanning 50 years, concerning the resolution of con‡icts using Bayesian heavy-tailed models. Con‡icts arise when di¤erent sources of information about the model parameters (e.g. prior information, or the information in individual observations) suggest quite di¤erent plausible regions for those parameters. Traditional Bayesian models based on normal distributions or other conjugate structures typically resolve con‡icts by centring the posterior at some… 

Figures from this paper

Robustness against conflicting prior information in regression
Including prior information about model parameters is a fundamental step of any Bayesian statistical analysis. It is viewed positively by some as it allows, among others, to quantitatively
• In Bayesian analysis, the prior distribution and the likelihood can conflict, in the sense that they can carry diverse information about the parameter of interest. The most common form of conflict
• In Bayesian analysis, the prior distribution and the likelihood can conflict, in the sense that they can carry diverse information about the parameter of interest. The most common form of conflict
Bayesian robustness modelling using the O‐regularly varying distributions
This work proposes a robustness modelling approach based on the O‐regularly varying class (ORV), which is a generalization of the regular variation family; however, the ORV class allows more flexible tails behaviour, which can improve the way in which the outlying information is discarded by the model.
Robust Bayesian Regression Analysis Using Ramsay-Novick Distributed Errors with Student-t Prior
This paper investigates bayesian treatment of regression modelling with Ramsay - Novick (RN)  distribution  specifically  developed for robust  inferential procedures. It falls into the category of
Default Bayesian Analysis with Global-Local Shrinkage Priors
The horseshoe priors, originally designed to handle sparsity, also possess regular variation and thus are appropriate for default Bayesian analysis, and are found to separate a low-dimensional signal embedded in high-dimensional noise, even for nonlinear functions.
Bayesian Restricted Likelihood Methods
Using several ii data analyses, the benefits of the new approach over the traditional approaches is made apparent and these benefits often manifest themselves through more realistic assumptions and more precise predictive performance.
A New Bayesian Approach to Robustness Against Outliers in Linear Regression
This paper proposes a model with super heavy-tailed errors, and proves that it is wholly robust, meaning that the impact of outliers gradually vanishes as they move further and further away form the general trend.
Robust Bayesian Reliability for Complex Systems under Prior-Data Conflict
  • G. Walter, F. Coolen
  • Computer Science
    ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering
  • 2018
An approach to mitigate prior-data conflict is presented, by considering sets of prior distributions to model limited knowledge on Weibull distributed component lifetimes, treating systems with arbitrary layout using the survival signature.
A Proposal for Informative Default Priors Scaled by the Standard Error of Estimates
Abstract If we have an unbiased estimate of some parameter of interest, then its absolute value is positively biased for the absolute value of the parameter. This bias is large when the


Bayesian robustness modeling using regularly varying distributions
Bayesian robustness modelling using heavy-tailed distributions provides a flexible approach to resolving problems of conflicts between the data and prior distributions. See Dawid (1973) and O’Hagan
Conflicting information and location parameter inference
It is shown that the posterior distribution converges in law to a density proportional to the product of the densities of the non-conflicting information, as the outliers (and/or the prior location) go to plus or minus infinity, at any given rate.
Bayesian Robustness Modelling of Location and Scale Parameters
Abstract.  The modelling process in Bayesian Statistics constitutes the fundamental stage of the analysis, since depending on the chosen probability laws the inferences may vary considerably. This is
Outlier Models and Prior Distributions in Bayesian Linear Regression
SUMMARY Bayesian inference in regression models is considered using heavy-tailed error distri- butions to accommodate outliers. The particular class of distributions that can be con- structed as
Outliers and Credence for Location Parameter Inference
Abstract Heavy-tailed distributions are important for modeling problems in which there may be outlying observations or parameters. Examples of their use are given in O'Hagan (1988). This article
On Robust Bayesian Analysis for Location and Scale Parameters
Dawid (1973,Biometrika60, 664?666) stated conditions in the univariate location model with known scale parameter needed for there to be either vanishing likelihood or prior influence on the posterior
A moment of indecision
We consider inference from a single observation X about a location parameter 0. That is, given 0 = 0, the density function of X has the form f (x -0) for some known function f. The prior distribution
On Outlier Rejection Phenomena in Bayes Inference
SUMMARY Inference is considered for a location parameter given a random sample. Outliers are not explicitly modelled, but rejection of extreme observations occurs naturally in any Bayesian analysis
Robust Sequential Approximate Bayesian Estimation
SUMMARY An approximation to the sequential updating of the distribution of location parameters of a linear time series model is developed for non-normal observations. The behaviour of the resulting
Bayesian Analysis of Linear and Non‐Linear Population Models by Using the Gibbs Sampler
Abstract : A fully Bayesian analysis of linear and nonlinear population models has previously been unavailable, as a consequence of the seeming impossibility of performing the necessary numerical