Posterior propriety in Bayesian extreme value analyses using reference priors

DOI10.5705/SS.2014.034MaRDI QIDQ2796878zbMATH OpenOpenAlexFDO

Authors Paul J. Northrop, Nicolas Attalides

Publication date 30 March 2016

Published in Statistica Sinica (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1505.04983

zbMATH Keywords

extreme value theorygeneralized extreme value distributiongeneralized Pareto distributionposterior proprietyreference prior

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Bayesian inference (62F15) Statistics of extreme values; tail inference (62G32)

Abstract: The Generalized Pareto (GP) and Generalized extreme value (GEV) distributions play an important role in extreme value analyses, as models for threshold excesses and block maxima respectively. For each of these distributions we consider Bayesian inference using "reference" prior distributions (in the general sense of priors constructed using formal rules) for the model parameters, specifically a Jeffreys prior, the maximal data information (MDI) prior and independent uniform priors on separate model parameters. We investigate the important issue of whether these improper priors lead to proper posterior distributions. We show that, in the GP and GEV cases, the MDI prior, unless modified, never yields a proper posterior and that in the GEV case this also applies to the Jeffreys prior. We also show that a sample size of three (four) is sufficient for independent uniform priors to yield a proper posterior distribution in the GP (GEV) case.

Recommendations

• Bayesian inference for extremes: accounting for the three extremal types
• Propriety of posterior distributions arising in categorical and survival models under generalized extreme value distribution
• scientific article; zbMATH DE number 1057894
• A matching prior for extreme quantile estimation of the generalized Pareto distribution
• A default Bayesian procedure for the generalized Pareto distribution

Cited in

(22)

• A Poisson process reparameterisation for Bayesian inference for extremes
• Consistency of Bayesian inference for multivariate max-stable distributions
• Posterior properties of the Weibull distribution for censored data
• Flexible link functions in nonparametric binary regression with Gaussian process priors
• A flexible cure rate model for spatially correlated survival data based on generalized extreme value distribution and Gaussian process priors
• Default priors based on pseudo-likelihoods for the Poisson-GPD model
• scientific article; zbMATH DE number 1057894 (Why is no real title available?)
• Full Bayesian significance test for extremal distributions
• Propriety of posterior distributions arising in categorical and survival models under generalized extreme value distribution
• Power Laws Distributions in Objective Priors
• Reference Priors for the Generalized Extreme Value Distribution
• A modeler's guide to extreme value software
• Flexible link functions in a joint model of binary and longitudinal data
• Empirical Bayes inference for the block maxima method
• Properties of k -record posteriors for the Weibull model
• On the posterior property of the Rician distribution
• Posterior properties with censored responses using the gamma distribution
• A utopic adventure in the modelling of conditional univariate and multivariate extremes. EVA (2023) Conference Data Challenge: team Yahabe
• On posterior properties of the two parameter gamma family of distributions
• Objective Bayesian inference for the Exponential-Logarithmic distribution
• Comparing different objective priors for extremes
• Estimation and uncertainty quantification for extreme quantile regions

This page was built for publication: Posterior propriety in Bayesian extreme value analyses using reference priors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2796878)