Servicenavigation


Sie sind hier:

Director

Walter Krämer          

+49 231 755 - 3125

walterk@statistik.tu-dortmund.de

 

Executive office, management

Thorsten Ziebach

+49 231 755 - 3122

thorsten.ziebach@udo.edu

 

An M-Estimator of the Multivariate Tail Dependence

The tail dependence structure is modeled by the stable tail dependence func-tion, which is assumed to be parametric. An M-estimator of the unknownparameter vector is de ned as the value of the parameter vector that mini-mizes the distance between a vector of weighted integrals of the tail depen-dence function and empirical counterparts of these integrals. Under minimalconditions, this minimization problem has with probability tending to one aunique, global solution. The estimator is consistent and asymptotically nor-mal. Its performance is illustrated on logistic and factor models. A variantof this estimator and its asymptotic properties for the case of meta-ellipticaldistributions are also presented.

 

An M-Estimator of the Multivariate Tail Dependence

Speaker: Andrea Krajina (Institut für Mathematische Stochastik, Universität Göttingen)

When and where?

Tuesday, May 24, 2011, 4.30 pm, M / E25