MULTIVARIATE EXCESS DISTRIBUTIONS UNDER ASYMPTOTIC INDEPENDENCE

 

  • Prof. Anne-Laure Fougères – Université Paul-Sabatier, Toulouse, France
  • FCUL (DEIO) – Campo Grande – Bloco C/6 – Piso 4 – Sala 6.4.30 -14:30
  • Quarta-feira, 2 de Março de 2005
 

Multivariate excess distributions under asymptotic independence

Estimating the probability for a vector (X1, …, Xd) to be in a set A, given that one component Xi is extreme, is a problem which has several applications, in hydrology for example, or in finance. Multivariate extreme value theory provides a useful answer as long as the vector considered satisfies the property of asymptotic dependence. In this talk, focus is done on asymptotic independent vectors, and a review of recent answers to the problem is provided.