Premium
A hierarchical Bayesian spatio‐temporal model for extreme precipitation events
Author(s) -
Ghosh Souparno,
Mallick Bani K.
Publication year - 2011
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/env.1043
Subject(s) - bayesian probability , hierarchy , markov chain monte carlo , computer science , data set , series (stratigraphy) , component (thermodynamics) , set (abstract data type) , precipitation , hierarchical database model , bayesian inference , data mining , statistics , artificial intelligence , mathematics , meteorology , geography , geology , paleontology , physics , economics , market economy , thermodynamics , programming language
We propose a new approach to model a sequence of spatially distributed time series of extreme values. Unlike common practice, we incorporate spatial dependence directly in the likelihood and allow the temporal component to be captured at the second level of hierarchy. Inferences about the parameters and spatio‐temporal predictions are obtained via MCMC technique. The model is fitted to a gridded precipitation data set collected over 99 years across the continental U.S. Copyright © 2010 John Wiley & Sons, Ltd.