Sojourn times and the fragility index

DOI10.1016/J.SPA.2011.11.009MaRDI QIDQ765893zbMATH OpenOpenAlexFDO

Authors Michael Falk, Martin Hofmann

Publication date 22 March 2012

Published in Stochastic Processes and their Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1107.5696

zbMATH Keywords

copula processexcursion timeexpected shortfallfragility indexfunctional domain of attractiongeneralized Pareto processmax-stable processsojourn time

Mathematics Subject Classification ID

Limit theorems in probability theory (60F99) Stationary stochastic processes (60G10) Extreme value theory; extremal stochastic processes (60G70)

Abstract: We investigate the sojourn time above a high threshold of a continuous stochastic process Y on [0,1]. It turns out that the limit, as the threshold increases, of the expected sojourn time given that it is positive, exists if the copula process corresponding to Y is in the functional domain of attraction of of an extreme value process. This limit coincides with the limit of the fragility index corresponding to finite (n-)dimensional distributions of Y as n and the threshold increase. If the process is in a certain neighborhood of a generalized Pareto process, then we can replace the constant threshold by a general threshold function and we can compute the asymptotic sojourn time distribution. An extreme value process is a prominent example. Given that there is an exceedance at some t_0 above the threshold, we can also compute the asymptotic distribution of the time cluster length, which the process spends above the threshold function.

Review text

Let \((Y_t)_{t\in[0,1]}\) be a stationary process with continuous sample paths and continuous marginal distribution function \(F\). The authors consider the sojourn time \(S(s)=\int_0^11_{\{Y_t>s\}}dt\) above a threshold \(s\) and its discretized version \(S_n(s)=n^{-1}\sum_{i=1}^n1_{\{Y_{i/n}>s\}}\). The conditional expectation \(FI_n(s)=E(S_n(s)\mid S_n(s)>0)\) is called fragility index and the authors explicitly compute and identify the limit \[ \lim_{n\to\infty}\lim_{s\uparrow\omega(F)}FI_n(s)=\lim_{s\uparrow\omega(F)}E(S(s)\mid S(s)>0) \] under the condition that the copula process \((F(Y_t))_{t\in[0,1]}\) belongs to the functional domain of attraction of a max-stable process, where \(\omega(F)=\sup\{x\in\mathbb R:\,F(x)<1\}\). The presented method further enables to explicitly compute the cumulative expected shortfall \[ ES(s)=E\left(\left.\int_0^1(Y_t-s)1_{\{Y_t>s\}}dt\right| S(s)>0\right). \] The limit result is extended to considering threshold functions instead of the constant threshold \(s\) in certain models for which the copula process is close to a generalized Pareto process. These considerations also enable the authors to compute the limit distribution of the excursion time above a threshold.

Recommendations

• Sojourn times of Gaussian processes with trend
• Asymptotic conditional distribution of exceedance counts: fragility index with different margins
• Sojourns of stationary processes in rare sets
• Asymptotic conditional distribution of exceedance counts
• Approximation of sojourn times of Gaussian processes

Cites work

• A multivariate piecing-together approach with an application to operational loss data
• Asymptotic conditional distribution of exceedance counts
• Extreme value theory. An introduction.
• scientific article; zbMATH DE number 107615 (Why is no real title available?)
• max-infinitely divisible and max-stable sample continuous processes
• On convergence toward an extreme value distribution in \(C[0,1]\)
• On max-stable processes and the functional D-norm
• On spatial extremes: with application to a rainfall problem
• On the excursion random measure of stationary processes
• Theory of Random Sets

Cited in

(2)

• Asymptotic conditional distribution of exceedance counts: fragility index with different margins
• Testing for a \(\delta \)-neighborhood of a generalized Pareto copula

This page was built for publication: Sojourn times and the fragility index

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q765893)