Title: A space-time process for extremes: Application to precipitation data
Authors: Gwladys Toulemonde - Université de Montpellier (France) [presenting]
Jean-Noel Bacro - Universite de Montpellier (France)
Carlo Gaetan - Ca Foscari University of Venice (Italy)
Thomas Opitz - BioSP, INRA (France)
Abstract: The statistical modeling of space-time extremes in environmental applications is a valuable approach to understand complex dependences in observed data and to generate realistic scenarios for impact models. Motivated by hourly rainfall data in Southern France presenting asymptotic independence, we propose a novel hierarchical model for high threshold exceedances defined over continuous space and time by embedding a space-time Gamma process convolution for the rate of an exponential variable, leading to asymptotic independence in space and time. This construction permits keeping marginal distributions which are coherent with univariate extreme value theory. Statistical inference is based on a pairwise likelihood for the observed censored excesses. The practical usefulness of our model is illustrated on the previously-mentioned hourly precipitation data set and comparisons with alternative censored Gaussian random fields are discussed.