Parametric and Nonparametric Estimators for Bivariate Extremes

• Activity Number: 497
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: IMS
• Author(s): Sabrina Vettori*+ and Raphael Huser and Marc G. Genton
• Companies: and and King Abdullah University of Science and Technology
• Keywords: extreme value analysis ; bivariate extremes ; logistic model ; threshold-based exceedances

Abstract

The practical implementation of extremes values analysis is of great importance in many disciplines. In a wide range of applications it is of interest to model not only the sizes of extremes, but also the joint behavior of several variables at high levels, such as natural phenomena at distinct locations. Several parametric and non parametric methods have been developed to model the extremal dependence between two or more variables of interest. Our goal is to investigate, through a simulation study, the performance of several threshold-based or block maximum estimators under different dependence scenarios in the bivariate case, focusing on the comparison between parametric and non parametric approaches.