B0644
Title: Nonparametric extreme quantile estimation for spatial data
Authors: Aladji Bassene - University of Lille (France)
Aliou Diop - University of Gaston Berger (Senegal)
Baba Thiam - LEM Universite Lille 3 (France)
Sophie Dabo - University of Lille (France) [presenting]
Abstract: Spatial statistics includes any techniques which study phenomenons observed on spatial sets. Such phenomenons appear in a variety of fields: epidemiology, environmental science, econometrics, image processing and many others. Complex issues arise in spatial analysis, many of which are neither clearly defined nor completely resolved, and form the basis for current researches. This is the case for instance in statistics of extremes, where data are often spatial, and so spatial location can acts as a surrogate for risk factors. More recently, there has been increased interest in non-parametric statistical models for spatial extremes. We are interested in quantile estimation for heavy tailed models when data are available in space. More precisely, we consider a non-parametric conditional quantile estimate where the explanatory and response variables are real-valued random fields. The asymptotic distribution of the proposed estimator is established under some mixing condition. The skills of the methods are illustrated on simulations.
Journal