Bernoulli and tail-dependence compatibility

DOI10.1214/15-AAP1128zbMATH Open1385.60029arXiv1606.08212WikidataQ85715902 ScholiaQ85715902MaRDI QIDQ303963FDOQ303963

Authors: Paul Embrechts, Marius Hofert, Ruodu Wang 

Publication date: 23 August 2016

Published in: The Annals of Applied Probability (Search for Journal in Brave)

Abstract: The tail-dependence compatibility problem is introduced. It raises the question whether a given $dimesd$-matrix of entries in the unit interval is the matrix of pairwise tail-dependence coefficients of a $d$-dimensional random vector. The problem is studied together with Bernoulli-compatible matrices, that is, matrices which are expectations of outer products of random vectors with Bernoulli margins. We show that a square matrix with diagonal entries being 1 is a tail-dependence matrix if and only if it is a Bernoulli-compatible matrix multiplied by a constant. We introduce new copula models to construct tail-dependence matrices, including commonly used matrices in statistics.

Full work available at URL: https://arxiv.org/abs/1606.08212

Recommendations

• A note on dependence modeling for Bernoulli variables
• Bernoulli randomness and Bernoulli normality
• Asymptotics for dependent Bernoulli random variables
• On Bernoulli distributions
• Limit behaviors for dependent Bernoulli variables
• Limit theorems for dependent Bernoulli variables with statistical inference
• scientific article; zbMATH DE number 6862726
• On the almost sure invariance principle for dependent Bernoulli random variables
• Tail dependence from a distributional point of view
• Some limit theorems for dependent Bernoulli random variables

zbMATH Keywords

tail dependencecopulasmatricescompatibilityBernoulli random vectorsinsurance application

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Exact distribution theory in statistics (62E15) Measures of association (correlation, canonical correlation, etc.) (62H20)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Extreme value properties of multivariate \(t\) copulas
• Construction of asymmetric multivariate copulas
• Title not available (Why is that?)
• Structured credit portfolio analysis, baskets \& CDOs
• Asymptotic results for the sum of dependent non-identically distributed random variables
• An Introduction to Credit Risk Modeling
• Characterization of dependence concepts in normal distributions
• The safest dependence structure among risks.
• Tail correlation functions of max-stable processes
• The realization problem for tail correlation functions
• Range of correlation matrices for dependent Bernoulli random variables
• A class of multivariate copulas with bivariate Fréchet marginal copulas

Cited In (25)

• Observations on industry practice in the construction of large correlation structures for risk and capital margins
• \(t\)-copula from the viewpoint of tail dependence matrices
• Tail-dependence, exceedance sets, and metric embeddings
• Copula-based measures of asymmetry between the lower and upper tail probabilities
• Distributional compatibility for change of measures
• A framework for measuring association of random vectors via collapsed random variables
• Calibrating Distribution Models from PELVE
• COMPATIBILITY AND ATTAINABILITY OF MATRICES OF CORRELATION-BASED MEASURES OF CONCORDANCE
• Geometry of discrete copulas
• Title not available (Why is that?)
• Invariant correlation under marginal transforms
• On attainability of Kendall's tau matrices and concordance signatures
• Matrix compatibility and correlation mixture representation of generalized Gini's gamma
• On tail dependence matrices. The realization problem for parametric families
• Admissible Bernoulli correlations
• Copula modeling from Abe Sklar to the present day
• Membership testing for Bernoulli and tail-dependence matrices
• Asymptotic independence and support detection techniques for heavy-tailed multivariate data
• Compatible matrices of Spearman's rank correlation
• On structure, family and parameter estimation of hierarchical Archimedean copulas
• A study of one-factor copula models from a tail dependence perspective
• The realization problem for tail correlation functions
• Extremes and regular variation
• Measuring non-exchangeable tail dependence using tail copulas
• Behaviour of multivariate tail dependence coefficients

Uses Software

• QRM

This page was built for publication: Bernoulli and tail-dependence compatibility

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q303963)