Exponentiated generalized Pareto distribution: Properties and applications towards extreme value theory

Seyoon Lee, Joseph H.T. Kim

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The GPD is a central distribution in modelling heavy tails in many applications. Applying the GPD to actual datasets however is not trivial. In this paper we propose the Exponentiated GPD (exGPD), created via log-transform of the GPD variable, which has less sample variability. Various distributional quantities of the exGPD are derived analytically. As an application we also propose a new plot based on the exGPD as an alternative to the Hill plot to identify the tail index of heavy tailed datasets, and carry out simulation studies to compare the two.

Original languageEnglish
Pages (from-to)2014-2038
Number of pages25
JournalCommunications in Statistics - Theory and Methods
Volume48
Issue number8
DOIs
Publication statusPublished - 2019 Apr 18

Fingerprint

Generalized Pareto Distribution
Extreme Value Theory
Tail Index
Heavy Tails
Trivial
Simulation Study
Transform
Alternatives
Modeling

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{3e91658bf76d4d6dbfdcfca76c95ca03,
title = "Exponentiated generalized Pareto distribution: Properties and applications towards extreme value theory",
abstract = "The GPD is a central distribution in modelling heavy tails in many applications. Applying the GPD to actual datasets however is not trivial. In this paper we propose the Exponentiated GPD (exGPD), created via log-transform of the GPD variable, which has less sample variability. Various distributional quantities of the exGPD are derived analytically. As an application we also propose a new plot based on the exGPD as an alternative to the Hill plot to identify the tail index of heavy tailed datasets, and carry out simulation studies to compare the two.",
author = "Seyoon Lee and Kim, {Joseph H.T.}",
year = "2019",
month = "4",
day = "18",
doi = "10.1080/03610926.2018.1441418",
language = "English",
volume = "48",
pages = "2014--2038",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "8",

}

TY - JOUR

T1 - Exponentiated generalized Pareto distribution

T2 - Properties and applications towards extreme value theory

AU - Lee, Seyoon

AU - Kim, Joseph H.T.

PY - 2019/4/18

Y1 - 2019/4/18

N2 - The GPD is a central distribution in modelling heavy tails in many applications. Applying the GPD to actual datasets however is not trivial. In this paper we propose the Exponentiated GPD (exGPD), created via log-transform of the GPD variable, which has less sample variability. Various distributional quantities of the exGPD are derived analytically. As an application we also propose a new plot based on the exGPD as an alternative to the Hill plot to identify the tail index of heavy tailed datasets, and carry out simulation studies to compare the two.

AB - The GPD is a central distribution in modelling heavy tails in many applications. Applying the GPD to actual datasets however is not trivial. In this paper we propose the Exponentiated GPD (exGPD), created via log-transform of the GPD variable, which has less sample variability. Various distributional quantities of the exGPD are derived analytically. As an application we also propose a new plot based on the exGPD as an alternative to the Hill plot to identify the tail index of heavy tailed datasets, and carry out simulation studies to compare the two.

UR - http://www.scopus.com/inward/record.url?scp=85043345075&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043345075&partnerID=8YFLogxK

U2 - 10.1080/03610926.2018.1441418

DO - 10.1080/03610926.2018.1441418

M3 - Article

AN - SCOPUS:85043345075

VL - 48

SP - 2014

EP - 2038

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 8

ER -