Space-time trends detection and dependence modelling approaches by functional Peaks-Over-Thresholds. Application to precipitation in Burkina Faso
Résumé
In this paper, we propose a method for estimating trends in extreme spatiotemporal processes using both information from marginal distributions and dependence structure. We combine two statistical approaches of extreme value theory allowing on the one hand to handle temporal and spatial non-stationarities via a tail trend function with a spatio-temporal structure in the marginal distributions and by modelling on the other hand the dependence structure by a latent stationary process using generalized-Pareto processes. This methodology for trend analysis of extreme events is applied to precipitation data from Burkina Faso. We show that a significant increasing trend for the 50 and 100 years return levels in some parts of the country.
Origine : Fichiers produits par l'(les) auteur(s)