An M-estimator of Spatial Tail Dependence

Tail dependence models for distributions attracted to a max-stable law are tted using observations above a high threshold. To cope with spatial, high-dimensional data, a rankbased M-estimator is proposed relying on bivariate margins only. A data-driven weight matrix is used to minimize the asymptotic variance. Empirical process arguments show that the estimator is consistent and asymptotically normal. Its nite-sample performance is assessed in simulation experiments involving popular max-stable processes perturbed with additive noise. An analysis of wind speed data from the Netherlands illustrates the method.