Home > Research > Publications & Outputs > Effects of mis-specification in bivariate extre...
View graph of relations

Effects of mis-specification in bivariate extreme value problems.

Research output: Contribution to journalJournal article

Published

Standard

Effects of mis-specification in bivariate extreme value problems. / Dupuis, Debbie J.; Tawn, Jonathan A.

In: Extremes, Vol. 4, No. 4, 12.2001, p. 315-330.

Research output: Contribution to journalJournal article

Harvard

APA

Vancouver

Author

Dupuis, Debbie J. ; Tawn, Jonathan A. / Effects of mis-specification in bivariate extreme value problems. In: Extremes. 2001 ; Vol. 4, No. 4. pp. 315-330.

Bibtex

@article{6e641bbf0aef4a33aa15890c9ab796f7,
title = "Effects of mis-specification in bivariate extreme value problems.",
abstract = "The need to incorporate the structure of complex problems in extreme value analyzes, and the requirement to exploit all the limited information that is available, has led to the increased use of advanced dependence models. When they are appropriate, these dependence models can lead to substantial benefits over simpler univariate extreme value methods. Here we explore some inference problems for the marginal and conditional distributions caused by model mis-specification. We find distinct differences in estimation characteristics when the dependence structure is asymptotically dependent or asymptotically independent, and that conditional models can be substantially improved if the variables are standardized to have common marginal distributions.",
keywords = "asymptotic independence - bivariate extreme value distribution - conditional distribution - covariates - dependence - generalized extreme-value distribution",
author = "Dupuis, {Debbie J.} and Tawn, {Jonathan A.}",
year = "2001",
month = "12",
doi = "10.1023/A:1016540012032",
language = "English",
volume = "4",
pages = "315--330",
journal = "Extremes",
issn = "1386-1999",
publisher = "Springer Netherlands",
number = "4",

}

RIS

TY - JOUR

T1 - Effects of mis-specification in bivariate extreme value problems.

AU - Dupuis, Debbie J.

AU - Tawn, Jonathan A.

PY - 2001/12

Y1 - 2001/12

N2 - The need to incorporate the structure of complex problems in extreme value analyzes, and the requirement to exploit all the limited information that is available, has led to the increased use of advanced dependence models. When they are appropriate, these dependence models can lead to substantial benefits over simpler univariate extreme value methods. Here we explore some inference problems for the marginal and conditional distributions caused by model mis-specification. We find distinct differences in estimation characteristics when the dependence structure is asymptotically dependent or asymptotically independent, and that conditional models can be substantially improved if the variables are standardized to have common marginal distributions.

AB - The need to incorporate the structure of complex problems in extreme value analyzes, and the requirement to exploit all the limited information that is available, has led to the increased use of advanced dependence models. When they are appropriate, these dependence models can lead to substantial benefits over simpler univariate extreme value methods. Here we explore some inference problems for the marginal and conditional distributions caused by model mis-specification. We find distinct differences in estimation characteristics when the dependence structure is asymptotically dependent or asymptotically independent, and that conditional models can be substantially improved if the variables are standardized to have common marginal distributions.

KW - asymptotic independence - bivariate extreme value distribution - conditional distribution - covariates - dependence - generalized extreme-value distribution

U2 - 10.1023/A:1016540012032

DO - 10.1023/A:1016540012032

M3 - Journal article

VL - 4

SP - 315

EP - 330

JO - Extremes

JF - Extremes

SN - 1386-1999

IS - 4

ER -