Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas

Tim Bedford, Alireza Daneshkhah, Kevin Wilson

Research output: Contribution to journalArticle

19 Citations (Scopus)
6 Downloads (Pure)

Abstract

Many applications of risk analysis require us to jointly model multiple uncertain quantities. Bayesian networks and copulas are two common approaches to modeling joint uncertainties with probability distributions. This article focuses on new methodologies for copulas by developing work of Cooke, Bedford, Kurowica, and others on vines as a way of constructing higher dimensional distributions that do not suffer from some of the restrictions of alternatives such as the multivariate Gaussian copula. The article provides a fundamental approximation result, demonstrating that we can approximate any density as closely as we like using vines. It further operationalizes this result by showing how minimum information copulas can be used to provide parametric classes of copulas that have such good levels of approximation. We extend previous approaches using vines by considering nonconstant conditional dependencies, which are particularly relevant in financial risk modeling. We discuss how such models may be quantified, in terms of expert judgment or by fitting data, and illustrate the approach by modeling two financial data sets.
Original languageEnglish
Pages (from-to)792-815
Number of pages23
JournalRisk Analysis
Volume36
Issue number4
DOIs
Publication statusPublished - 2 Sep 2015
Externally publishedYes

Fingerprint

Risk analysis
Uncertainty
Bayesian networks
Probability distributions
Joints
Dependency (Psychology)
Datasets

Bibliographical note

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.

Keywords

  • copula
  • Entropy
  • risk modelling
  • Pair-copula Construction
  • Information

Cite this

Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas. / Bedford, Tim ; Daneshkhah, Alireza; Wilson, Kevin.

In: Risk Analysis, Vol. 36, No. 4, 02.09.2015, p. 792-815.

Research output: Contribution to journalArticle

Bedford, Tim ; Daneshkhah, Alireza ; Wilson, Kevin. / Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas. In: Risk Analysis. 2015 ; Vol. 36, No. 4. pp. 792-815.
@article{163f67ce9b8c4be5b687b6c4e4fb086f,
title = "Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas",
abstract = "Many applications of risk analysis require us to jointly model multiple uncertain quantities. Bayesian networks and copulas are two common approaches to modeling joint uncertainties with probability distributions. This article focuses on new methodologies for copulas by developing work of Cooke, Bedford, Kurowica, and others on vines as a way of constructing higher dimensional distributions that do not suffer from some of the restrictions of alternatives such as the multivariate Gaussian copula. The article provides a fundamental approximation result, demonstrating that we can approximate any density as closely as we like using vines. It further operationalizes this result by showing how minimum information copulas can be used to provide parametric classes of copulas that have such good levels of approximation. We extend previous approaches using vines by considering nonconstant conditional dependencies, which are particularly relevant in financial risk modeling. We discuss how such models may be quantified, in terms of expert judgment or by fitting data, and illustrate the approach by modeling two financial data sets.",
keywords = "copula, Entropy, risk modelling, Pair-copula Construction, Information",
author = "Tim Bedford and Alireza Daneshkhah and Kevin Wilson",
note = "This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.",
year = "2015",
month = "9",
day = "2",
doi = "10.1111/risa.12471",
language = "English",
volume = "36",
pages = "792--815",
journal = "Risk Analysis",
issn = "0272-4332",
publisher = "Wiley",
number = "4",

}

TY - JOUR

T1 - Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas

AU - Bedford, Tim

AU - Daneshkhah, Alireza

AU - Wilson, Kevin

N1 - This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

PY - 2015/9/2

Y1 - 2015/9/2

N2 - Many applications of risk analysis require us to jointly model multiple uncertain quantities. Bayesian networks and copulas are two common approaches to modeling joint uncertainties with probability distributions. This article focuses on new methodologies for copulas by developing work of Cooke, Bedford, Kurowica, and others on vines as a way of constructing higher dimensional distributions that do not suffer from some of the restrictions of alternatives such as the multivariate Gaussian copula. The article provides a fundamental approximation result, demonstrating that we can approximate any density as closely as we like using vines. It further operationalizes this result by showing how minimum information copulas can be used to provide parametric classes of copulas that have such good levels of approximation. We extend previous approaches using vines by considering nonconstant conditional dependencies, which are particularly relevant in financial risk modeling. We discuss how such models may be quantified, in terms of expert judgment or by fitting data, and illustrate the approach by modeling two financial data sets.

AB - Many applications of risk analysis require us to jointly model multiple uncertain quantities. Bayesian networks and copulas are two common approaches to modeling joint uncertainties with probability distributions. This article focuses on new methodologies for copulas by developing work of Cooke, Bedford, Kurowica, and others on vines as a way of constructing higher dimensional distributions that do not suffer from some of the restrictions of alternatives such as the multivariate Gaussian copula. The article provides a fundamental approximation result, demonstrating that we can approximate any density as closely as we like using vines. It further operationalizes this result by showing how minimum information copulas can be used to provide parametric classes of copulas that have such good levels of approximation. We extend previous approaches using vines by considering nonconstant conditional dependencies, which are particularly relevant in financial risk modeling. We discuss how such models may be quantified, in terms of expert judgment or by fitting data, and illustrate the approach by modeling two financial data sets.

KW - copula

KW - Entropy

KW - risk modelling

KW - Pair-copula Construction

KW - Information

U2 - 10.1111/risa.12471

DO - 10.1111/risa.12471

M3 - Article

VL - 36

SP - 792

EP - 815

JO - Risk Analysis

JF - Risk Analysis

SN - 0272-4332

IS - 4

ER -