Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 380 - Advances in Bayesian Extreme Value Analysis
Type: Topic Contributed
Date/Time: Wednesday, August 10, 2022 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #322987
Title: A Flexible Bayesian Hierarchical Modeling Framework for Spatially Dependent Peaks-Over-Threshold Data
Author(s): Rishikesh Yadav* and Raphael Huser and Thomas Opitz
Companies: King Abdullah University of Science and Technology and King Abdullah University of Science and Technology (KAUST) and INRAe
Keywords: Bayesian hierarchical modeling ; extreme event; stochastic gradient Langevin dynamics; sub-asymptotic modeling; threshold exceedance.

In this work, we develop a constructive modeling framework for extreme spatial threshold exceedances, based on general product mixtures of random fields possessing light or heavy-tailed margins and various spatial dependence characteristics, which are suitably designed to provide high flexibility in the tail and at pre-asymptotic levels. Our proposed model is akin to a recently proposed Gamma-Gamma model, but it possesses a higher degree of flexibility in its joint tail structure, capturing strong dependence more easily. Thanks to the model’s hierarchical formulation, the inference may be conveniently performed based on Markov chain Monte Carlo (MCMC) methods, and we demonstrate how the stochastic gradient Langevin dynamics (SGLD) algorithm can be exploited to fit our model very efficiently in relatively high spatio-temporal dimensions, while simultaneously censoring non-threshold exceedances and performing spatial prediction at multiple sites. We explore the theoretical properties of the proposed model, and illustrate the proposed methodology by simulation and application to daily precipitation data from Eastern Spain measured at about 100 stations over the period 2011-2020.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program