Main Article Content

Nonparametric condence intervals for tail dependence based on copulas


Cheikh Tidiane Seck
Diam Ba
Gane Samb Lo

Abstract

Abstract. We propose nonparametric asymptotic condence intervals for the upper and lower tail dependence coecients. These latter are obtained from condence bands established for the copula function itself and based upon three kernel-type estimators. We show the performance of these condence intervals through a simulation study. We also apply these results to nancial data stemming from the CAC 40 stock index which reveals the existence of extreme dependence between larger values of the opening and closing prices for this index during the considered period.

Resume. Ce papier presente des intervalles de conance asymptotiques pour les coe- cients de dependance de queue inferieure et superieure. Ces derniers sont obtenus de facon nonparametrique a partir de bandes de conance presque sures obtenues pour la fonction copule elle meme et basees sur des estimateurs a noyau. Nous montrons ensuite la performance de ces intervalles de conance a travers une etude de simulation. Une application de ces resultats sur des donnees nancieres issues de l'indice boursier CAC 40 revelent une dependance extreme entre les valeurs d'ouverture et de fermeture de cet indice durant la periode etudiee.

Key words: Tail dependence coecient, Condence intervals, Kernel estimators, Copula
function.


Journal Identifiers


eISSN:
print ISSN: 2316-090X