Corrected inference about the extreme Expected Shortfall in the general Max-Domain of Attraction
- 27 June 2025
- journal article
- Published by Oxford University Press (OUP) in Information and Inference: A Journal of the IMA
- Vol. 14 (3)
Abstract
The use of the Expected Shortfall as a solution for various deficiencies of quantiles has gained substantial traction in the field of risk assessment over the last 20 years. Existing approaches to its inference at extreme levels remain limited to distributions that are both heavy-tailed and have a finite second tail moment. This constitutes a strong restriction in areas like finance and environmental science, where the random variable of interest may have a much heavier tail or, at the opposite, may be light-tailed or short-tailed. Under a wider semiparametric extreme value framework, we develop comprehensive asymptotic theory for Expected Shortfall estimation above extreme quantiles in the class of distributions with finite first tail moment, regardless of whether the underlying extreme value index is positive, negative or zero. By relying on the moment estimators of the scale and shape extreme value parameters, we construct corrected asymptotic confidence intervals whose finite-sample coverage is found to be close to the nominal level on simulated data. We illustrate the usefulness of our construction on two sets of financial loss returns and flood insurance claims data.Keywords
All Related Versions
Funding Information
- French National Research Agency (ANR-19-CE40-0013, ANR-23-CE40-0009, ANR-17-EURE-0010, ANR-11-LABX-0020-01)
- TSE-HEC ACPR Chair
- Chair Stress Test
- RISK Management and Financial Steering of the Foundation Ecole Polytechnique
This publication has 34 references indexed in Scilit:
- Simple sufficient criteria for second-order extended regular variation in the Gumbel domain of attraction: The case of Weibull-tailed distributionsExtremes, 2025
- Bias-reduced and variance-corrected asymptotic Gaussian inference about extreme expectilesStatistics and Computing, 2024
- Extreme expectile estimation for short-tailed dataJournal of Econometrics, 2024
- Extreme-value based estimation of the conditional tail moment with application to reinsurance ratingInsurance: Mathematics and Economics, 2022
- On automatic bias reduction for extreme expectile estimationStatistics and Computing, 2022
- Quantile Function Expansion Using Regularly Varying FunctionsMethodology and Computing in Applied Probability, 2017
- Bias correction in extreme value statistics with index around zeroExtremes, 2012
- Laws of Small Numbers: Extremes and Rare Events2011
- Extreme Value Theory2006
- Statistics of Extremes2004