Non-parametric estimation of extreme risk measures from conditional heavy-tailed distributions

Abstract : In this paper, we introduce a new risk measure, the so-called Conditional Tail Moment. It is the moment of order a>0 of the loss distribution above the upper alpha-quantile. Estimating the Conditional Tail Moment permits to estimate all risk measures based on conditional moments such as Conditional Tail Expectation, Conditional Value-at-Risk or Conditional Tail Variance. Here, we focus on the estimation of these risk measures in case of extreme losses (where alpha converges to 0). It is moreover assumed that the loss distribution is heavy-tailed and depends on a covariate. The estimation method thus combines nonparametric kernel methods with extreme-value statistics. The asymptotic distribution of the estimators is established and their finite sample behavior is illustrated both on simulated data and on a real data set of daily rainfalls in the Cévennes-Vivarais region (France).
Type de document :
Article dans une revue
Scandinavian Journal of Statistics, Wiley, 2014, 41 (4), pp.988-1012. <10.1111/sjos.12078>
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00830647
Contributeur : Stephane Girard <>
Soumis le : jeudi 28 novembre 2013 - 11:08:09
Dernière modification le : jeudi 2 février 2017 - 21:16:13
Document(s) archivé(s) le : lundi 3 mars 2014 - 16:16:19

Fichier

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

Identifiants

Collections

Citation

Jonathan El Methni, Laurent Gardes, Stephane Girard. Non-parametric estimation of extreme risk measures from conditional heavy-tailed distributions. Scandinavian Journal of Statistics, Wiley, 2014, 41 (4), pp.988-1012. <10.1111/sjos.12078>. <hal-00830647v5>

Partager

Métriques

Consultations de
la notice

867

Téléchargements du document

531