[Statlist] Research Seminar in Statistics | *FRIDAY 22 MARCH 2024* | GSEM, University of Geneva

gsem-support-instituts g@em-@upport-|n@t|tut@ @end|ng |rom un|ge@ch
Mon Mar 18 10:58:18 CET 2024


Dear all,

We are pleased to invite you to our next Research Seminar, organized by Professor Sebastian Engelke on behalf of the Research Center for Statistics
< https://www.unige.ch/gsem/en/research/institutes/rcs/team/ >.

FRIDAY 22 MARCH 2024, at 11:15 am, Uni Mail M 4220.

Stochastic Geometry and Extrapolation of Spatial Extremes
(jointly with Elena de Bernardino, Ryan Cotsakis, and Anne Estrade)
Thomas OPITZ, Biostatistics and Spatial Processes, INRAE Avignon, France
< https://biosp.mathnum.inrae.fr/homepage-thomas-opitz >

ABSTRACT:
Extreme events in climatic and other environmental processes often show spatial dependence. Geometric properties of exceedance regions above high thresholds, such as their area, perimeter or number of connected components, provide relevant information about the spatial structure of extremes and specifically about how their dependence strength changes at increasingly high thresholds.

The first part of this talk concerns the expected values of certain geometric properties in location- or scale-mixtures of a Gaussian process, where the mean or the standard deviation, respectively, of a stationary Gaussian process is a random variable. The results include exact formulas for peaks-over-threshold-stable limit processes (so-called Pareto processes) arising from the use of Gaussian or log-Gaussian spectral functions in the spectral construction of max-stable processes.

A second part of the talk concerns the notion of extremal range. Conditioned on exceedance of a high threshold at a location s, the extremal range at s is the random variable defined as the smallest distance from s to a location where there is a non-exceedance. It captures the rate at which the spatial extent of conditional extreme events scales for increasingly high thresholds. We study its distributional properties, propose parametric models and predict the median extremal range at extreme threshold levels.

An application of these geometric concepts to two large gridded daily temperature datasets, namely reanalyses and climate-model simulations for France, highlights decreasing extremal dependence for increasing threshold levels and also reveals substantial differences in joint tail decay rates between reanalyses and simulations.

> View the Research Seminar agenda: < https://www.unige.ch/gsem/en/research/seminars/rcs/ >

Regards,


Marie-Madeleine

Marie-Madeleine Novo
Assistant to the Research Institutes
gsem-support-instituts using unige.ch



More information about the Statlist mailing list