Tail Risk in the Tail: Estimating High Quantiles When a Related Variable is Extreme

• 成果类型: Article; Early Access
• 署名作者: Nolde, Natalia; Zhou, Chen; Zhou, Menglin
• 署名单位: University of British Columbia; Erasmus University Rotterdam - Excl Erasmus MC; Erasmus University Rotterdam
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459; 1537-274X
• DOI: 10.1080/01621459.2026.2639148
• 发表日期: 2026-05-19
• 关键词: CoVaR; Extreme conditional quantile estimation; Heavy Tails; Multivariate extreme value theory; Tail dependence function; dependence; INDEPENDENCE

摘要

In this article we address the problem of high quantile estimation conditional on a related variable being extreme. The problem set-up is of interest in a number applications to evaluate tail risk of a focal variable in the tail of a conditioning variable. A primary example we consider is the assessment of systemic risk in financial markets using a risk measure known as the conditional value-at-risk (CoVaR). The proposed estimator is based on a novel approach to handle the bivariate tail dependence structure through an adjustment factor that can be used in conjunction with univariate high quantile estimation techniques. We establish the asymptotic behavior of the estimator under relatively weak assumptions, and illustrate its performance via simulation studies and a real data example. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.