An Entropic View of Pickands' Theorem

Abstract : It is shown that distributions arising in Renyi-Tsallis maximum entropy setting are related to the generalized Pareto distributions (GPD) that are widely used for modeling the tails of distributions. The relevance of such modelization, as well as the ubiquity of GPD in practical situations follows from Balkema-De Haan-Pickands theorem on the distribution of excesses (over a high threshold). We provide an entropic view of this result, by showing that the distribution of a suitably normalized excess variable converges to the solution of a maximum Tsallis entropy, which is the GPD. This result resembles the entropic approach to the central limit theorem; however, the convergence in entropy proved here is weaker than the convergence in supremum norm given by Pickandspsila theorem.
Type de document :
Communication dans un congrès
International Symposium on Information Theory (ISIT 2008), Jul 2008, Toronto, Canada. Proceedings of the IEEE 2008 International Symposium on Information Theory, pp.2625-2628, 2008, <10.1109/ISIT.2008.4595467>
Liste complète des métadonnées


https://hal-upec-upem.archives-ouvertes.fr/hal-00621924
Contributeur : Jean-François Bercher <>
Soumis le : jeudi 17 mars 2016 - 11:51:02
Dernière modification le : mercredi 4 janvier 2017 - 16:19:25
Document(s) archivé(s) le : dimanche 13 novembre 2016 - 20:36:52

Fichier

isit2008.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Jean-François Bercher, Christophe Vignat. An Entropic View of Pickands' Theorem. International Symposium on Information Theory (ISIT 2008), Jul 2008, Toronto, Canada. Proceedings of the IEEE 2008 International Symposium on Information Theory, pp.2625-2628, 2008, <10.1109/ISIT.2008.4595467>. <hal-00621924>

Partager

Métriques

Consultations de
la notice

144

Téléchargements du document

40