ExpectHill estimation, extreme risk and heavy tails

Abstract : Risk measures of a financial position are traditionally based on quantiles. Replacing quantiles with their least squares analogues, called expectiles, has recently received increasing attention. The novel expectile-based risk measures satisfy all coherence requirements. We revisit their extreme value estimation for heavy-tailed distributions. First, we estimate the underlying tail index via weighted combinations of top order statistics and asymmetric least squares estimates. The resulting expectHill estimators are then used as the basis for estimating tail expectiles and Expected Shortfall. The asymptotic theory of the proposed estimators is provided, along with numerical simulations and applications to actuarial and financial data.
Document type :
Preprints, Working Papers, ...
Liste complète des métadonnées

Cited literature [27 references]  Display  Hide  Download

https://hal.inria.fr/hal-01856212
Contributor : Stephane Girard <>
Submitted on : Friday, August 10, 2018 - 10:08:47 AM
Last modification on : Friday, April 19, 2019 - 4:54:52 PM
Document(s) archivé(s) le : Sunday, November 11, 2018 - 12:20:24 PM

Files

DGS_XES_JE_main_final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01856212, version 1

Collections

Citation

Abdelaati Daouia, Stéphane Girard, Gilles Stupfler. ExpectHill estimation, extreme risk and heavy tails. 2018. ⟨hal-01856212⟩

Share

Metrics

Record views

380

Files downloads

79