Estimating extreme tail risk measures with generalized Pareto distribution

Myung Hyun Park, Joseph H.T. Kim

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The generalized Pareto distribution (GPD) has been widely used in modelling heavy tail phenomena in many applications. The standard practice is to fit the tail region of the dataset to the GPD separately, a framework known as the peaks-over-threshold (POT) in the extreme value literature. In this paper we propose a new GPD parameter estimator, under the POT framework, to estimate common tail risk measures, the Value-at-Risk (VaR) and Conditional Tail Expectation (also known as Tail-VaR) for heavy-tailed losses. The proposed estimator is based on a nonlinear weighted least squares method that minimizes the sum of squared deviations between the empirical distribution function and the theoretical GPD for the data exceeding the tail threshold. The proposed method properly addresses a caveat of a similar estimator previously advocated, and further improves the performance by introducing appropriate weights in the optimization procedure. Using various simulation studies and a realistic heavy-tailed model, we compare alternative estimators and show that the new estimator is highly competitive, especially when the tail risk measures are concerned with extreme confidence levels.

Original languageEnglish
Pages (from-to)91-104
Number of pages14
JournalComputational Statistics and Data Analysis
Volume98
DOIs
Publication statusPublished - 2016 Jun

Fingerprint

Generalized Pareto Distribution
Risk Measures
Tail
Extremes
Estimator
Peaks over Threshold
Value at Risk
Distribution functions
Empirical Distribution Function
Heavy Tails
Nonlinear Least Squares
Weighted Least Squares
Confidence Level
Extreme Values
Least Square Method
Deviation
Simulation Study
Minimise
Optimization
Alternatives

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

@article{a2de33f628aa451abfa7ce52f1e51d06,
title = "Estimating extreme tail risk measures with generalized Pareto distribution",
abstract = "The generalized Pareto distribution (GPD) has been widely used in modelling heavy tail phenomena in many applications. The standard practice is to fit the tail region of the dataset to the GPD separately, a framework known as the peaks-over-threshold (POT) in the extreme value literature. In this paper we propose a new GPD parameter estimator, under the POT framework, to estimate common tail risk measures, the Value-at-Risk (VaR) and Conditional Tail Expectation (also known as Tail-VaR) for heavy-tailed losses. The proposed estimator is based on a nonlinear weighted least squares method that minimizes the sum of squared deviations between the empirical distribution function and the theoretical GPD for the data exceeding the tail threshold. The proposed method properly addresses a caveat of a similar estimator previously advocated, and further improves the performance by introducing appropriate weights in the optimization procedure. Using various simulation studies and a realistic heavy-tailed model, we compare alternative estimators and show that the new estimator is highly competitive, especially when the tail risk measures are concerned with extreme confidence levels.",
author = "Park, {Myung Hyun} and Kim, {Joseph H.T.}",
year = "2016",
month = "6",
doi = "10.1016/j.csda.2015.12.008",
language = "English",
volume = "98",
pages = "91--104",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",

}

Estimating extreme tail risk measures with generalized Pareto distribution. / Park, Myung Hyun; Kim, Joseph H.T.

In: Computational Statistics and Data Analysis, Vol. 98, 06.2016, p. 91-104.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Estimating extreme tail risk measures with generalized Pareto distribution

AU - Park, Myung Hyun

AU - Kim, Joseph H.T.

PY - 2016/6

Y1 - 2016/6

N2 - The generalized Pareto distribution (GPD) has been widely used in modelling heavy tail phenomena in many applications. The standard practice is to fit the tail region of the dataset to the GPD separately, a framework known as the peaks-over-threshold (POT) in the extreme value literature. In this paper we propose a new GPD parameter estimator, under the POT framework, to estimate common tail risk measures, the Value-at-Risk (VaR) and Conditional Tail Expectation (also known as Tail-VaR) for heavy-tailed losses. The proposed estimator is based on a nonlinear weighted least squares method that minimizes the sum of squared deviations between the empirical distribution function and the theoretical GPD for the data exceeding the tail threshold. The proposed method properly addresses a caveat of a similar estimator previously advocated, and further improves the performance by introducing appropriate weights in the optimization procedure. Using various simulation studies and a realistic heavy-tailed model, we compare alternative estimators and show that the new estimator is highly competitive, especially when the tail risk measures are concerned with extreme confidence levels.

AB - The generalized Pareto distribution (GPD) has been widely used in modelling heavy tail phenomena in many applications. The standard practice is to fit the tail region of the dataset to the GPD separately, a framework known as the peaks-over-threshold (POT) in the extreme value literature. In this paper we propose a new GPD parameter estimator, under the POT framework, to estimate common tail risk measures, the Value-at-Risk (VaR) and Conditional Tail Expectation (also known as Tail-VaR) for heavy-tailed losses. The proposed estimator is based on a nonlinear weighted least squares method that minimizes the sum of squared deviations between the empirical distribution function and the theoretical GPD for the data exceeding the tail threshold. The proposed method properly addresses a caveat of a similar estimator previously advocated, and further improves the performance by introducing appropriate weights in the optimization procedure. Using various simulation studies and a realistic heavy-tailed model, we compare alternative estimators and show that the new estimator is highly competitive, especially when the tail risk measures are concerned with extreme confidence levels.

UR - http://www.scopus.com/inward/record.url?scp=84956883023&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84956883023&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2015.12.008

DO - 10.1016/j.csda.2015.12.008

M3 - Article

AN - SCOPUS:84956883023

VL - 98

SP - 91

EP - 104

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

ER -