Home > Research > Publications & Outputs > X-vine models for multivariate extremes

Electronic data

  • main-JRSSB

    Accepted author manuscript, 1.23 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Text available via DOI:

View graph of relations

X-vine models for multivariate extremes

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

X-vine models for multivariate extremes. / Kiriliouk, Anna; Lee, Jeongjin; Segers, Johan.
In: Journal of the Royal Statistical Society Series B: Statistical Methodology, 21.11.2024.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Kiriliouk, A, Lee, J & Segers, J 2024, 'X-vine models for multivariate extremes', Journal of the Royal Statistical Society Series B: Statistical Methodology. https://doi.org/10.1093/jrsssb/qkae105

APA

Kiriliouk, A., Lee, J., & Segers, J. (2024). X-vine models for multivariate extremes. Journal of the Royal Statistical Society Series B: Statistical Methodology. Advance online publication. https://doi.org/10.1093/jrsssb/qkae105

Vancouver

Kiriliouk A, Lee J, Segers J. X-vine models for multivariate extremes. Journal of the Royal Statistical Society Series B: Statistical Methodology. 2024 Nov 21. Epub 2024 Nov 21. doi: 10.1093/jrsssb/qkae105

Author

Kiriliouk, Anna ; Lee, Jeongjin ; Segers, Johan. / X-vine models for multivariate extremes. In: Journal of the Royal Statistical Society Series B: Statistical Methodology. 2024.

Bibtex

@article{2d3efaa2dd9544f9bc2550721df05b77,
title = "X-vine models for multivariate extremes",
abstract = "Regular vine sequences permit the organization of variables in a random vector along a sequence of trees. Vine-based dependence models have become greatly popular as a way to combine arbitrary bivariate copulas into higher-dimensional ones, offering flexibility, parsimony, and tractability. In this project, we use regular vine sequences to decompose and construct the exponent measure density of a multivariate extreme value distribution, or, equivalently, the tail copula density. Although these densities pose theoretical challenges due to their infinite mass, their homogeneity property offers simplifications. The theory sheds new light on existing parametric families and facilitates the construction of new ones, called X-vines. Computations proceed via recursive formulas in terms of bivariate model components. We develop simulation algorithms for X-vine multivariate Pareto distributions as well as methods for parameter estimation and model selection on the basis of threshold exceedances. The methods are illustrated by Monte Carlo experiments and a case study on US flight delay data.",
author = "Anna Kiriliouk and Jeongjin Lee and Johan Segers",
year = "2024",
month = nov,
day = "21",
doi = "10.1093/jrsssb/qkae105",
language = "English",
journal = "Journal of the Royal Statistical Society Series B: Statistical Methodology",

}

RIS

TY - JOUR

T1 - X-vine models for multivariate extremes

AU - Kiriliouk, Anna

AU - Lee, Jeongjin

AU - Segers, Johan

PY - 2024/11/21

Y1 - 2024/11/21

N2 - Regular vine sequences permit the organization of variables in a random vector along a sequence of trees. Vine-based dependence models have become greatly popular as a way to combine arbitrary bivariate copulas into higher-dimensional ones, offering flexibility, parsimony, and tractability. In this project, we use regular vine sequences to decompose and construct the exponent measure density of a multivariate extreme value distribution, or, equivalently, the tail copula density. Although these densities pose theoretical challenges due to their infinite mass, their homogeneity property offers simplifications. The theory sheds new light on existing parametric families and facilitates the construction of new ones, called X-vines. Computations proceed via recursive formulas in terms of bivariate model components. We develop simulation algorithms for X-vine multivariate Pareto distributions as well as methods for parameter estimation and model selection on the basis of threshold exceedances. The methods are illustrated by Monte Carlo experiments and a case study on US flight delay data.

AB - Regular vine sequences permit the organization of variables in a random vector along a sequence of trees. Vine-based dependence models have become greatly popular as a way to combine arbitrary bivariate copulas into higher-dimensional ones, offering flexibility, parsimony, and tractability. In this project, we use regular vine sequences to decompose and construct the exponent measure density of a multivariate extreme value distribution, or, equivalently, the tail copula density. Although these densities pose theoretical challenges due to their infinite mass, their homogeneity property offers simplifications. The theory sheds new light on existing parametric families and facilitates the construction of new ones, called X-vines. Computations proceed via recursive formulas in terms of bivariate model components. We develop simulation algorithms for X-vine multivariate Pareto distributions as well as methods for parameter estimation and model selection on the basis of threshold exceedances. The methods are illustrated by Monte Carlo experiments and a case study on US flight delay data.

U2 - 10.1093/jrsssb/qkae105

DO - 10.1093/jrsssb/qkae105

M3 - Journal article

JO - Journal of the Royal Statistical Society Series B: Statistical Methodology

JF - Journal of the Royal Statistical Society Series B: Statistical Methodology

ER -