Testing conditions and estimating parameters in extreme value theory: Application to environmental data

Helena Penalva, Dora Prata Gomes, M. Manuela Neves, Sandra Nunes

Research output: Contribution to journalArticle

1 Citation (Scopus)
2 Downloads (Pure)

Abstract

Extreme Value Theory has been asserting itself as one of the most important statistical theories for the applied sciences providing a solid theoretical basis for deriving statistical models describing extreme or even rare events. The efficiency of the inference and estimation procedures depends on the tail shape of the distribution underlying the data. In this work we will present a review of tests for assessing extreme value conditions and for the choice of the extreme value domain. Motivated by two real environmental problems we will apply those tests showing the need of performing such tests for choosing the most appropriate parameter estimation methods.

Original languageEnglish
Pages (from-to)187-207
Number of pages21
JournalRevstat Statistical Journal
Volume17
Issue number2
Publication statusPublished - 1 Apr 2019

Fingerprint

Extreme Value Theory
Extreme Values
Testing
Rare Events
Statistical Model
Parameter Estimation
Tail
Extremes

Keywords

  • Environmental data
  • Extreme values
  • Heavy-tailed distributions
  • Semi-parametric estimation
  • Statistical testing

Cite this

@article{e3b0478115c243b1a22cc768dc044071,
title = "Testing conditions and estimating parameters in extreme value theory: Application to environmental data",
abstract = "Extreme Value Theory has been asserting itself as one of the most important statistical theories for the applied sciences providing a solid theoretical basis for deriving statistical models describing extreme or even rare events. The efficiency of the inference and estimation procedures depends on the tail shape of the distribution underlying the data. In this work we will present a review of tests for assessing extreme value conditions and for the choice of the extreme value domain. Motivated by two real environmental problems we will apply those tests showing the need of performing such tests for choosing the most appropriate parameter estimation methods.",
keywords = "Environmental data, Extreme values, Heavy-tailed distributions, Semi-parametric estimation, Statistical testing",
author = "Helena Penalva and Gomes, {Dora Prata} and Neves, {M. Manuela} and Sandra Nunes",
note = "info:eu-repo/grantAgreement/FCT/5876/147410/PT# UID/MAT/00006/2019 (CEAUL) and UID/MAT/0297/2013 (CMA/UNL).",
year = "2019",
month = "4",
day = "1",
language = "English",
volume = "17",
pages = "187--207",
journal = "REVSTAT - STATISTICAL JOURNAL",
issn = "1645-6726",
publisher = "INST NACIONAL ESTATISTICA-INE",
number = "2",

}

Testing conditions and estimating parameters in extreme value theory: Application to environmental data. / Penalva, Helena; Gomes, Dora Prata; Neves, M. Manuela; Nunes, Sandra.

In: Revstat Statistical Journal, Vol. 17, No. 2, 01.04.2019, p. 187-207.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Testing conditions and estimating parameters in extreme value theory: Application to environmental data

AU - Penalva, Helena

AU - Gomes, Dora Prata

AU - Neves, M. Manuela

AU - Nunes, Sandra

N1 - info:eu-repo/grantAgreement/FCT/5876/147410/PT# UID/MAT/00006/2019 (CEAUL) and UID/MAT/0297/2013 (CMA/UNL).

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Extreme Value Theory has been asserting itself as one of the most important statistical theories for the applied sciences providing a solid theoretical basis for deriving statistical models describing extreme or even rare events. The efficiency of the inference and estimation procedures depends on the tail shape of the distribution underlying the data. In this work we will present a review of tests for assessing extreme value conditions and for the choice of the extreme value domain. Motivated by two real environmental problems we will apply those tests showing the need of performing such tests for choosing the most appropriate parameter estimation methods.

AB - Extreme Value Theory has been asserting itself as one of the most important statistical theories for the applied sciences providing a solid theoretical basis for deriving statistical models describing extreme or even rare events. The efficiency of the inference and estimation procedures depends on the tail shape of the distribution underlying the data. In this work we will present a review of tests for assessing extreme value conditions and for the choice of the extreme value domain. Motivated by two real environmental problems we will apply those tests showing the need of performing such tests for choosing the most appropriate parameter estimation methods.

KW - Environmental data

KW - Extreme values

KW - Heavy-tailed distributions

KW - Semi-parametric estimation

KW - Statistical testing

UR - http://www.scopus.com/inward/record.url?scp=85070689380&partnerID=8YFLogxK

M3 - Article

VL - 17

SP - 187

EP - 207

JO - REVSTAT - STATISTICAL JOURNAL

JF - REVSTAT - STATISTICAL JOURNAL

SN - 1645-6726

IS - 2

ER -