Statistics of heteroscedastic extremes
Einmahl,John de Haan,L.F.M. Zhou,C.
Einmahl,John
de Haan,L.F.M.
Zhou,C.
Abstract
We extend classical extreme value theory to non-identically distributed observations. When the tails of the distribution are proportional much of extreme value statistics remains valid. The proportionality function for the tails can be estimated non-parametrically along with the (common) extreme value index. For a positive extreme value index, joint asymptotic normality of both estimators is shown; they are asymptotically independent. We also establish asymptotic normality of a forecasted high quantile and develop tests for the proportionality function and for the validity of the model. We show through simulations the good performance of the procedures and also present an application to stock market returns. A main tool is the weak convergence of a weighted sequential tail empirical process.
Description
Date
2016-01
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Keywords
Extreme value statistics, functional limit theorems, non-identical distributions, sequential tail empirical process
Citation
Einmahl, J, de Haan, L F M & Zhou, C 2016, 'Statistics of heteroscedastic extremes', Journal of the Royal Statistical Society Series B-Statistical Methodology, vol. 78, no. 1, pp. 31-51. https://doi.org/10.1111/rssb.12099