Melina Mailhot (Concordia university, Montreal, Canada)
In this presentation, we will propose a smooth copula-based Generalized Extreme Value (GEV) model to map and predict extreme rainfall in Central Eastern Canada. The considered data contains a large portion of missing values, and one observes several non-concomitant record periods at different stations.
The proposed two-steps approach combines GEV parameters' smooth functions in space through the use of spatial covariates and a flexible hierarchical copula-based model to take into account dependence between the recording stations. The hierarchical copula structure is detected via a clustering algorithm implemented with an adapted version of the copula-based dissimilarity measure.