An exceptional max-stable process fully parameterized by its extremal coefficients

DOI10.3150/13-BEJ567zbMATH Open1323.60075arXiv1504.03459OpenAlexW1964733624MaRDI QIDQ2345121FDOQ2345121

Authors: Kirstin Strokorb, M. Schlather 

Publication date: 19 May 2015

Published in: Bernoulli (Search for Journal in Brave)

Abstract: The extremal coefficient function (ECF) of a max-stable process $X$ on some index set $T$ assigns to each finite subset $AsubsetT$ the effective number of independent random variables among the collection ${X_t}_{tinA}$. We introduce the class of Tawn-Molchanov processes that is in a 1:1 correspondence with the class of ECFs, thus also proving a complete characterization of the ECF in terms of negative definiteness. The corresponding Tawn-Molchanov process turns out to be exceptional among all max-stable processes sharing the same ECF in that its dependency set is maximal w.r.t. inclusion. This entails sharp lower bounds for the finite dimensional distributions of arbitrary max-stable processes in terms of its ECF. A spectral representation of the Tawn-Molchanov process and stochastic continuity are discussed. We also show how to build new valid ECFs from given ECFs by means of Bernstein functions.

Full work available at URL: https://arxiv.org/abs/1504.03459

Recommendations

• Characterization and construction of max-stable processes
• Tail correlation functions of max-stable processes
• Capturing the multivariate extremal index: bounds and interconnections
• Spatial extremes and max-stable processes
• On approximating max-stable processes and constructing extremal copula functions

zbMATH Keywords

max-stable processesextremal coefficientsBerstein functionsTawn-Molchanov process

Mathematics Subject Classification ID

Extreme value theory; extremal stochastic processes (60G70)

Cites Work

• Maxima of normal random vectors: Between independence and complete dependence
• Modelling pairwise dependence of maxima in space
• Statistics of Extremes
• Stationary max-stable fields associated to negative definite functions
• A spectral representation for max-stable processes
• Models for stationary max-stable random fields
• Inequalities for the extremal coefficients of multivariate extreme value distributions
• On spatial extremes: with application to a rainfall problem
• On the likelihood function of Gaussian max-stable processes
• On the structure and representations of max-stable processes
• Variograms for spatial max-stable random fields
• Title not available (Why is that?)
• Title not available (Why is that?)
• Theory of Random Sets
• Spatial modeling of extreme snow depth
• Convex geometry of max-stable distributions
• Dependence measures for extreme value analyses
• Title not available (Why is that?)
• Title not available (Why is that?)
• The extremogram: a correlogram for extreme events
• max-infinitely divisible and max-stable sample continuous processes
• Limit theory for multivariate sample extremes
• Characterization theorems for the Gneiting class of space-time covariances
• Spectral representations of sum- and max-stable processes
• Extremal stochastic integrals: a parallel between max-stable processes and \(\alpha\)-stable processes
• On the ergodicity and mixing of max-stable processes
• Functional limit theorems for sums of independent geometric Lévy processes
• From Schoenberg to Pick-Nevanlinna: toward a complete picture of the variogram class
• Regularity of the sample paths of a general second order random field
• On the generation of a multivariate extreme value distribution with prescribed tail dependence parameter matrix

Cited In (15)

• Tail-dependence, exceedance sets, and metric embeddings
• Stochastic ordering in multivariate extremes
• Max-stable random sup-measures with comonotonic tail dependence
• On approximating max-stable processes and constructing extremal copula functions
• Maxima of independent, non-identically distributed Gaussian vectors
• ABC model selection for spatial extremes models applied to south Australian maximum temperature data
• Upper bounds on value-at-risk for the maximum portfolio loss
• Distributionally robust inference for extreme value-at-risk
• On Extremal Index of max-stable stationary processes
• Characterization and construction of max-stable processes
• Tail correlation functions of max-stable processes
• The realization problem for tail correlation functions
• Extremal \(t\) processes: elliptical domain of attraction and a spectral representation
• Long range dependence for stable random processes
• Multidimensional extremal dependence coefficients

This page was built for publication: An exceptional max-stable process fully parameterized by its extremal coefficients

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2345121)