Robust bounds in multivariate extremes

TL;DR

This paper develops robust asymptotic bounds for multivariate extreme value probabilities that remain reliable under model misspecification, with applications to financial risk assessment.

Contribution

It introduces a method to derive explicit, robust bounds on exceedance probabilities in multivariate extremes, accounting for dependence model uncertainty.

Findings

Robust bounds outperform classical bounds under model misspecification.

Explicit bounds are simple to compute and applicable in real-world scenarios.

Application to financial portfolios demonstrates practical utility.

Abstract

Extreme value theory provides an asymptotically justified framework for estimation of exceedance probabilities in regions where few or no observations are available. For multivariate tail estimation, the strength of extremal dependence is crucial and it is typically modeled by a parametric family of spectral distributions. In this work we provide asymptotic bounds on exceedance probabilities that are robust against misspecification of the extremal dependence model. They arise from optimizing the statistic of interest over all dependence models within some neighborhood of the reference model. A certain relaxation of these bounds yields surprisingly simple and explicit expressions, which we propose to use in applications. We show the effectiveness of the robust approach compared to classical confidence bounds when the model is misspecified. The results are further applied to quantify the effect of model uncertainty on the Value-at-Risk of a financial portfolio.