Assessing models for estimation and methods for uncertainty quantification for spatial return levels

Yi Cao, Bo Li

Research output: Contribution to journalArticle

Abstract

The return level estimation is an essential topic in studying spatial extremes for environmental data. Recently, various models for spatial extremes have emerged, which generally yield different estimates for return levels, given the same data. In the meantime, several approaches that obtain confidence intervals (CIs) for return levels have arisen, and the results from different approaches can also largely disagree. These pose natural questions for assessing different return level estimation methods and different CI derivation approaches. In this article, we compare an array of popular models for spatial extremes in return level estimation, as well as three approaches in CI derivation, through extensive Monte Carlo simulations. Our results show that in general, max-stable models yield return level estimates with similar mean squared error, and the spatial generalized extreme value model also provides comparable estimates. The bootstrap method is recommended for max-stable models to compute the CI, and the profile likelihood CI works well for spatial generalized extreme value. We also evaluate the methods for return level interpolation at unknown spatial locations and find that kriging of marginal return level estimates performs as well as max-stable models.

Original languageEnglish (US)
Article numbere2508
JournalEnvironmetrics
Volume30
Issue number2
DOIs
StatePublished - Mar 1 2019

Fingerprint

Uncertainty Quantification
confidence interval
Confidence interval
Stable Models
Extremes
Extreme Values
Estimate
Model
Profile Likelihood
bootstrapping
Bootstrap Method
Kriging
estimation method
kriging
Mean Squared Error
interpolation
method
Monte Carlo Simulation
Interpolate
Unknown

Keywords

  • generalized extreme value
  • return level estimation
  • spatial extremes
  • uncertainty quantification

ASJC Scopus subject areas

  • Statistics and Probability
  • Ecological Modeling

Cite this

Assessing models for estimation and methods for uncertainty quantification for spatial return levels. / Cao, Yi; Li, Bo.

In: Environmetrics, Vol. 30, No. 2, e2508, 01.03.2019.

Research output: Contribution to journalArticle

@article{e4e595e293b14370806fbec93da90051,
title = "Assessing models for estimation and methods for uncertainty quantification for spatial return levels",
abstract = "The return level estimation is an essential topic in studying spatial extremes for environmental data. Recently, various models for spatial extremes have emerged, which generally yield different estimates for return levels, given the same data. In the meantime, several approaches that obtain confidence intervals (CIs) for return levels have arisen, and the results from different approaches can also largely disagree. These pose natural questions for assessing different return level estimation methods and different CI derivation approaches. In this article, we compare an array of popular models for spatial extremes in return level estimation, as well as three approaches in CI derivation, through extensive Monte Carlo simulations. Our results show that in general, max-stable models yield return level estimates with similar mean squared error, and the spatial generalized extreme value model also provides comparable estimates. The bootstrap method is recommended for max-stable models to compute the CI, and the profile likelihood CI works well for spatial generalized extreme value. We also evaluate the methods for return level interpolation at unknown spatial locations and find that kriging of marginal return level estimates performs as well as max-stable models.",
keywords = "generalized extreme value, return level estimation, spatial extremes, uncertainty quantification",
author = "Yi Cao and Bo Li",
year = "2019",
month = "3",
day = "1",
doi = "10.1002/env.2508",
language = "English (US)",
volume = "30",
journal = "Environmetrics",
issn = "1180-4009",
publisher = "John Wiley and Sons Ltd",
number = "2",

}

TY - JOUR

T1 - Assessing models for estimation and methods for uncertainty quantification for spatial return levels

AU - Cao, Yi

AU - Li, Bo

PY - 2019/3/1

Y1 - 2019/3/1

N2 - The return level estimation is an essential topic in studying spatial extremes for environmental data. Recently, various models for spatial extremes have emerged, which generally yield different estimates for return levels, given the same data. In the meantime, several approaches that obtain confidence intervals (CIs) for return levels have arisen, and the results from different approaches can also largely disagree. These pose natural questions for assessing different return level estimation methods and different CI derivation approaches. In this article, we compare an array of popular models for spatial extremes in return level estimation, as well as three approaches in CI derivation, through extensive Monte Carlo simulations. Our results show that in general, max-stable models yield return level estimates with similar mean squared error, and the spatial generalized extreme value model also provides comparable estimates. The bootstrap method is recommended for max-stable models to compute the CI, and the profile likelihood CI works well for spatial generalized extreme value. We also evaluate the methods for return level interpolation at unknown spatial locations and find that kriging of marginal return level estimates performs as well as max-stable models.

AB - The return level estimation is an essential topic in studying spatial extremes for environmental data. Recently, various models for spatial extremes have emerged, which generally yield different estimates for return levels, given the same data. In the meantime, several approaches that obtain confidence intervals (CIs) for return levels have arisen, and the results from different approaches can also largely disagree. These pose natural questions for assessing different return level estimation methods and different CI derivation approaches. In this article, we compare an array of popular models for spatial extremes in return level estimation, as well as three approaches in CI derivation, through extensive Monte Carlo simulations. Our results show that in general, max-stable models yield return level estimates with similar mean squared error, and the spatial generalized extreme value model also provides comparable estimates. The bootstrap method is recommended for max-stable models to compute the CI, and the profile likelihood CI works well for spatial generalized extreme value. We also evaluate the methods for return level interpolation at unknown spatial locations and find that kriging of marginal return level estimates performs as well as max-stable models.

KW - generalized extreme value

KW - return level estimation

KW - spatial extremes

KW - uncertainty quantification

UR - http://www.scopus.com/inward/record.url?scp=85050509053&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050509053&partnerID=8YFLogxK

U2 - 10.1002/env.2508

DO - 10.1002/env.2508

M3 - Article

VL - 30

JO - Environmetrics

JF - Environmetrics

SN - 1180-4009

IS - 2

M1 - e2508

ER -