A comparison of spatial extreme value models. Application to precipitation data.

Quentin Sebille (1) , Anne-Laure Fougères (1) , Cécile Mercadier (1)

1 PSPM - Probabilités, statistique, physique mathématique

In this paper, focus is done on spatial models for extreme events and on their respective efficiency regarding the estimation of two risk measures: one extrapolating marginal distributions and one summarizing the spatial bivariate dependence of extremes. A wide comparison is performed on a simulation plan that has been specifically designed from a daily precipitation data set. The objective of this paper is twofold: firstly, pointing out the inherent properties of each model, and secondly, advising users on how to choose the model depending on the specific type of risk.

Mots clés: max-stable processes; Spatial modelling of extreme events; precipitation data; extreme value theory; hierarchical models; spatial prediction

Domaines: Statistiques [stat]

Quentin Sebille, Anne-Laure Fougères, Cécile Mercadier. A comparison of spatial extreme value models. Application to precipitation data.. 2016. ⟨hal-01300751⟩