[Statlist] Research Seminar in Statistics *FRIDAY 23 MARCH 2018* GSEM, University of Geneva

[gsem-support-instituts]

Dear All,

We are pleased to invite you to our next Research Seminar.

Looking forward to seeing you

Organizers :                                                                                   
E. Cantoni - S. Engelke - D. La Vecchia - E. Ronchetti
S. Sperlich - F. Trojani - M.-P. Victoria-Feser

FRIDAY 23 MARCH 2018 at 11:15am, Uni-Mail M 5220

The new family of Fisher copulas to model upper tail dependence and radial asymmetry: properties and application to high-dimensional rainfall data
Anne-Catherine FAVRE - Institut national polytechnique de Grenoble and Ecole nationale supérieure de l'Energie, l'Eau et l'Environnement, and Laboratoire d'étude des Transferts en Hydrologie et Environnement, Grenoble, France

ABSTRACT:
Joint precipitation data measured at a large number of stations typically show tail asymmetry and significant upper tail dependence. Unfortunately, many multivariate dependence models that are commonly used in large dimensions like the Normal and the Student copulas are radially symmetric, while the recently introduced chi-square copula is asymmetric, but its tail dependence coefficients are null. In order to circumvent the limitations of the available models, the new family of Fisher copulas is introduced; it is shown that these dependence models are tail asymmetric and allow for upper tail dependence, among other characteristics. Two semi-parametric strategies for parameter estimation in this class of copulas are proposed and their efficiency in small and moderate sample sizes is investigated with the help of simulations. The usefulness of the parametric Fisher copula family is then illustrated on the modeling of precipitation data observed at 105 stations within or close to the Aare river catchment in Switzerland.


Visit the website: https://www.unige.ch/gsem/en/research/seminars/rcs/