A continuous updating weighted least squares estimator of tail dependence in high dimensions

John H.J. Einmahl, Anna Kiriliouk, Johan Segers

Research output: Contribution to journalArticle

Abstract

Likelihood-based procedures are a common way to estimate tail dependence parameters. They are not applicable, however, in non-differentiable models such as those arising from recent max-linear structural equation models. Moreover, they can be hard to compute in higher dimensions. An adaptive weighted least-squares procedure matching nonparametric estimates of the stable tail dependence function with the corresponding values of a parametrically specified proposal yields a novel minimum-distance estimator. The estimator is easy to calculate and applies to a wide range of sampling schemes and tail dependence models. In large samples, it is asymptotically normal with an explicit and estimable covariance matrix. The minimum distance obtained forms the basis of a goodness-of-fit statistic whose asymptotic distribution is chi-square. Extensive Monte Carlo simulations confirm the excellent finite-sample performance of the estimator and demonstrate that it is a strong competitor to currently available methods. The estimator is then applied to disentangle sources of tail dependence in European stock markets.

LanguageEnglish
Pages205-233
Number of pages29
JournalExtremes
Volume21
Issue number2
DOIs
Publication statusPublished - 1 Jun 2018
Externally publishedYes

Fingerprint

Weighted Least Squares Estimator
Tail Dependence
Higher Dimensions
Updating
Estimator
Minimum Distance Estimator
Dependence Function
Covariance matrix
Structural Equation Model
Chi-square
Weighted Least Squares
Minimum Distance
Goodness of fit
Stock Market
Statistics
Estimate
Sampling
Asymptotic distribution
Statistic
Linear equation

Keywords

  • Brown–Resnick process
  • Extremal coefficient
  • Max-linear model
  • Multivariate extremes
  • Stable tail dependence function

Cite this

Einmahl, John H.J. ; Kiriliouk, Anna ; Segers, Johan. / A continuous updating weighted least squares estimator of tail dependence in high dimensions. In: Extremes. 2018 ; Vol. 21, No. 2. pp. 205-233.
@article{1421c43d16ca4ab3b5027be24bd5eaaa,
title = "A continuous updating weighted least squares estimator of tail dependence in high dimensions",
abstract = "Likelihood-based procedures are a common way to estimate tail dependence parameters. They are not applicable, however, in non-differentiable models such as those arising from recent max-linear structural equation models. Moreover, they can be hard to compute in higher dimensions. An adaptive weighted least-squares procedure matching nonparametric estimates of the stable tail dependence function with the corresponding values of a parametrically specified proposal yields a novel minimum-distance estimator. The estimator is easy to calculate and applies to a wide range of sampling schemes and tail dependence models. In large samples, it is asymptotically normal with an explicit and estimable covariance matrix. The minimum distance obtained forms the basis of a goodness-of-fit statistic whose asymptotic distribution is chi-square. Extensive Monte Carlo simulations confirm the excellent finite-sample performance of the estimator and demonstrate that it is a strong competitor to currently available methods. The estimator is then applied to disentangle sources of tail dependence in European stock markets.",
keywords = "Brown–Resnick process, Extremal coefficient, Max-linear model, Multivariate extremes, Stable tail dependence function",
author = "Einmahl, {John H.J.} and Anna Kiriliouk and Johan Segers",
year = "2018",
month = "6",
day = "1",
doi = "10.1007/s10687-017-0303-7",
language = "English",
volume = "21",
pages = "205--233",
journal = "Extremes",
issn = "1386-1999",
publisher = "Springer New York",
number = "2",

}

A continuous updating weighted least squares estimator of tail dependence in high dimensions. / Einmahl, John H.J.; Kiriliouk, Anna; Segers, Johan.

In: Extremes, Vol. 21, No. 2, 01.06.2018, p. 205-233.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A continuous updating weighted least squares estimator of tail dependence in high dimensions

AU - Einmahl, John H.J.

AU - Kiriliouk, Anna

AU - Segers, Johan

PY - 2018/6/1

Y1 - 2018/6/1

N2 - Likelihood-based procedures are a common way to estimate tail dependence parameters. They are not applicable, however, in non-differentiable models such as those arising from recent max-linear structural equation models. Moreover, they can be hard to compute in higher dimensions. An adaptive weighted least-squares procedure matching nonparametric estimates of the stable tail dependence function with the corresponding values of a parametrically specified proposal yields a novel minimum-distance estimator. The estimator is easy to calculate and applies to a wide range of sampling schemes and tail dependence models. In large samples, it is asymptotically normal with an explicit and estimable covariance matrix. The minimum distance obtained forms the basis of a goodness-of-fit statistic whose asymptotic distribution is chi-square. Extensive Monte Carlo simulations confirm the excellent finite-sample performance of the estimator and demonstrate that it is a strong competitor to currently available methods. The estimator is then applied to disentangle sources of tail dependence in European stock markets.

AB - Likelihood-based procedures are a common way to estimate tail dependence parameters. They are not applicable, however, in non-differentiable models such as those arising from recent max-linear structural equation models. Moreover, they can be hard to compute in higher dimensions. An adaptive weighted least-squares procedure matching nonparametric estimates of the stable tail dependence function with the corresponding values of a parametrically specified proposal yields a novel minimum-distance estimator. The estimator is easy to calculate and applies to a wide range of sampling schemes and tail dependence models. In large samples, it is asymptotically normal with an explicit and estimable covariance matrix. The minimum distance obtained forms the basis of a goodness-of-fit statistic whose asymptotic distribution is chi-square. Extensive Monte Carlo simulations confirm the excellent finite-sample performance of the estimator and demonstrate that it is a strong competitor to currently available methods. The estimator is then applied to disentangle sources of tail dependence in European stock markets.

KW - Brown–Resnick process

KW - Extremal coefficient

KW - Max-linear model

KW - Multivariate extremes

KW - Stable tail dependence function

UR - http://www.scopus.com/inward/record.url?scp=85028769886&partnerID=8YFLogxK

U2 - 10.1007/s10687-017-0303-7

DO - 10.1007/s10687-017-0303-7

M3 - Article

VL - 21

SP - 205

EP - 233

JO - Extremes

T2 - Extremes

JF - Extremes

SN - 1386-1999

IS - 2

ER -