Title: Bayesian Hierarchical Models for Extreme Values Observed over Space and Time

Abstract: We propose a hierarchical Bayesian approach
for modeling a collection of spatially-referenced time series of
extreme values. We assume that the observations follow Generalized
Extreme Value(GEV) distributions whose locations and scales are
spatially dependent. Indeed, the GEV response surface can also be
smoothed through a transformed spatial Gaussian Process. The
models are fitted using a Markov Chain Monte Carlo (MCMC)
algorithm to enable inference for parameters and to provide
spatio-temporal predictions. Metropolis- adjusted Langevin (MALA)
algorithm is introduced to generate samples from the complex
multivariate posterior distributions. Computations in large sample
size spatial model will also be discussed.