Online Program Home
My Program

Abstract Details

Activity Number: 108 - Multivariate Extremes: Theory and Applications
Type: Invited
Date/Time: Monday, July 29, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Risk Analysis
Abstract #300290
Title: Semiparametric Estimation for Multivariate Extremes
Author(s): John P Nolan* and Anne-Laure Fougeres and Cecile Mercadier
Companies: American University and Univesrity of Lyon and University of Lyon
Keywords: extreme values; multivariate; semi-parametric; max-projections
Abstract:

Modeling multivariate extremes is an interesting problem because many problems involve multiple observations; it is a challenging problem because the joint distribution involves estimating a measure; an infinite dimensional problem. We present an approach to fitting such models based on two principles. First, classes of semi-parametric models are described that can approximate any extreme value distribution. Second, an estimation method based on max-projections is used to make the estimation problem tractable. An R package mvevd that implements this approach is described.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program