Non-parametric estimation of extreme risk measures from conditional heavy-tailed distributions - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Scandinavian Journal of Statistics Year : 2014

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).
Fichier principal
Vignette du fichier
CTMtmp.pdf (356.58 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00830647 , version 1 (05-06-2013)
hal-00830647 , version 2 (10-06-2013)
hal-00830647 , version 3 (22-08-2013)
hal-00830647 , version 4 (19-09-2013)
hal-00830647 , version 5 (28-11-2013)

Identifiers

Cite

Jonathan El Methni, Laurent Gardes, Stéphane Girard. Non-parametric estimation of extreme risk measures from conditional heavy-tailed distributions. Scandinavian Journal of Statistics, 2014, 41 (4), pp.988-1012. ⟨10.1111/sjos.12078⟩. ⟨hal-00830647v5⟩
1057 View
1293 Download

Altmetric

Share

Gmail Facebook X LinkedIn More