Extremal dependence of random scale constructions

DOI10.1007/S10687-019-00353-3MaRDI QIDQ2283053zbMATH OpenOpenAlexWikidataFDO

Authors Sebastian Engelke, T. Opitz, Jennifer L. Wadsworth

Publication date 27 December 2019

Published in Extremes (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/1803.04221

zbMATH Keywords

copulaextreme value theoryresidual tail dependencetail dependence

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Extreme value theory; extremal stochastic processes (60G70) Measures of association (correlation, canonical correlation, etc.) (62H20)

Abstract: A bivariate random vector can exhibit either asymptotic independence or dependence between the largest values of its components. When used as a statistical model for risk assessment in fields such as finance, insurance or meteorology, it is crucial to understand which of the two asymptotic regimes occurs. Motivated by their ubiquity and flexibility, we consider the extremal dependence properties of vectors with a random scale construction $(X_1,X_2)=R(W_1,W_2)$, with non-degenerate $R>0$ independent of $(W_1,W_2)$. Focusing on the presence and strength of asymptotic tail dependence, as expressed through commonly-used summary parameters, broad factors that affect the results are: the heaviness of the tails of $R$ and $(W_1,W_2)$, the shape of the support of $(W_1,W_2)$, and dependence between $(W_1,W_2)$. When $R$ is distinctly lighter tailed than $(W_1,W_2)$, the extremal dependence of $(X_1,X_2)$ is typically the same as that of $(W_1,W_2)$, whereas similar or heavier tails for $R$ compared to $(W_1,W_2)$ typically result in increased extremal dependence. Similar tail heavinesses represent the most interesting and technical cases, and we find both asymptotic independence and dependence of $(X_1,X_2)$ possible in such cases when $(W_1,W_2)$ exhibit asymptotic independence. The bivariate case often directly extends to higher-dimensional vectors and spatial processes, where the dependence is mainly analyzed in terms of summaries of bivariate sub-vectors. The results unify and extend many existing examples, and we use them to propose new models that encompass both dependence classes.

Recommendations

• Extreme residual dependence for random vectors and processes
• On extremal dependence of block vectors
• Samples with a limit shape, multivariate extremes, and risk
• The Extremal Dependence Measure and Asymptotic Independence
• Limit laws for random vectors with an extreme component

Cites work

• A Conditional Approach for Multivariate Extreme Values (with Discussion)
• Asymptotic independence for unimodal densities
• Asymptotics of random contractions
• Bivariate distributions with given extreme value attractor
• Convolution equivalence and distributions of random sums
• Convolution equivalence and infinite divisibility
• Convolution tails, product tails and domains of attraction
• Convolutions of Long-Tailed and Subexponential Distributions
• Extremal \(t\) processes: elliptical domain of attraction and a spectral representation
• Extremal attractors of Liouville copulas
• Extreme behaviour for bivariate elliptical distributions
• Extreme residual dependence for random vectors and processes
• Extreme value theory for space-time processes with heavy-tailed distributions
• Extremes of randomly scaled Gumbel risks
• Factor copula models for replicated spatial data
• Functional regular variations, Pareto processes and peaks over threshold
• Geometric interpretation of the residual dependence coefficient
• Heavy-Tail Phenomena
• High risk scenarios and extremes. A geometric approach
• scientific article; zbMATH DE number 4030574 (Why is no real title available?)
• scientific article; zbMATH DE number 469373 (Why is no real title available?)
• scientific article; zbMATH DE number 699423 (Why is no real title available?)
• scientific article; zbMATH DE number 1026574 (Why is no real title available?)
• scientific article; zbMATH DE number 1085999 (Why is no real title available?)
• scientific article; zbMATH DE number 775283 (Why is no real title available?)
• scientific article; zbMATH DE number 3238248 (Why is no real title available?)
• Limit conditional distributions for bivariate vectors with polar representation
• Limit laws for random vectors with an extreme component
• Modeling Spatial Processes with Unknown Extremal Dependence Class
• Modelling across extremal dependence classes
• Multivariate Archimax copulas
• Multivariate Archimedean copulas, \(d\)-monotone functions and \(\ell _{1}\)-norm symmetric distributions
• Multivariate extremes, aggregation and dependence in elliptical distributions
• Multivariate generalized Pareto distributions
• Multivariate peaks over thresholds models
• On conditional extreme values of random vectors with polar representation
• On multivariate Gaussian tails
• On the residual dependence index of elliptical distributions
• Statistics of Extremes
• Subexponentiality of the product of independent random variables
• Tail asymptotics of light-tailed Weibull-like sums
• The generalized Pareto process; with a view towards application and simulation

Cited in

(26)

• Rank-based estimation under asymptotic dependence and independence, with applications to spatial extremes
• On the tail behaviour of aggregated random variables
• Modeling spatial extremes using normal mean-variance mixtures
• Heavy-tailed phase-type distributions: a unified approach
• Asymmetric tail dependence modeling, with application to cryptocurrency market data
• Tail dependence functions of the bivariate Hüsler-Reiss model
• Semiparametric bivariate modelling with flexible extremal dependence
• Asymptotic dependence for light-tailed homothetic densities
• Linking representations for multivariate extremes via a limit set
• A Vecchia approximation for high-dimensional Gaussian cumulative distribution functions arising from spatial data
• Hierarchical Transformed Scale Mixtures for Flexible Modeling of Spatial Extremes on Datasets With Many Locations
• Total positivity in multivariate extremes
• An efficient workflow for modelling high-dimensional spatial extremes
• Advances in statistical modeling of spatial extremes
• Partial Tail-Correlation Coefficient Applied to Extremal-Network Learning
• Tail dependence coefficients of moving average processes driven by exponential-tailed Lévy noise
• Spatial modeling of extremes and an angular component
• Efficient Modeling of Spatial Extremes over Large Geographical Domains
• Graphical models for infinite measures with applications to extremes
• Fast spatial simulation of extreme high-resolution radar precipitation data using integrated nested Laplace approximations
• Non-stationary max-stable models with an application to heavy rainfall data
• Multivariate peaks-over-threshold with latent variable representations of generalized Pareto vectors
• Extremal graphical modeling with latent variables via convex optimization
• Clustered Archimax copulas
• Neural Bayes estimation and selection of complex bivariate extremal dependence models
• On the worst and least possible asymptotic dependence

This page was built for publication: Extremal dependence of random scale constructions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2283053)