Home > Research > Publications & Outputs > A penalised piecewise-linear model for non-stat...

Research

Associated organisational units

Links

Text available via DOI:

Keywords

View graph of relations

A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold. / Barlow, Anna Maria; Mackay, Ed; Eastoe, Emma et al.
In: Ocean Engineering, Vol. 267, 113265, 01.01.2023.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Barlow, AM, Mackay, E, Eastoe, E & Jonathan, P 2023, 'A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold', Ocean Engineering, vol. 267, 113265. https://doi.org/10.1016/j.oceaneng.2022.113265

APA

Barlow, A. M., Mackay, E., Eastoe, E., & Jonathan, P. (2023). A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold. Ocean Engineering, 267, Article 113265. https://doi.org/10.1016/j.oceaneng.2022.113265

Vancouver

Barlow AM, Mackay E, Eastoe E, Jonathan P. A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold. Ocean Engineering. 2023 Jan 1;267:113265. Epub 2022 Dec 2. doi: 10.1016/j.oceaneng.2022.113265

Author

Barlow, Anna Maria ; Mackay, Ed ; Eastoe, Emma et al. / A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold. In: Ocean Engineering. 2023 ; Vol. 267.

Bibtex

@article{769cf07b33d547ae8269dd721c355dba,
title = "A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold",
abstract = "Metocean extremes often vary systematically with covariates such as direction and season. In this work, we present non-stationary models for the size and rate of occurrence of peaks over threshold of metocean variables with respect to one- or two-dimensional covariates. The variation of model parameters with covariate is described using a piecewise-linear function in one or two dimensions, defined with respect to pre-specified node locations on the covariate domain. Parameter roughness is regulated to provide optimal predictive performance, assessed using cross-validation, within a penalised likelihood framework for inference. Parameter uncertainty is quantified using bootstrap resampling. The models are used to estimate extremes of storm-peak significant wave height with respect to direction and season for a site in the northern North Sea. A covariate representation based on a triangulation of the direction-season domain with six nodes gives good predictive performance. The penalised piecewise-linear framework provides a flexible representation of covariate effects at reasonable computational cost.",
keywords = "Extreme, Non-stationary, Covariate, Significant wave height, Penalised likelihood",
author = "Barlow, {Anna Maria} and Ed Mackay and Emma Eastoe and Philip Jonathan",
year = "2023",
month = jan,
day = "1",
doi = "10.1016/j.oceaneng.2022.113265",
language = "English",
volume = "267",
journal = "Ocean Engineering",
issn = "0029-8018",
publisher = "Elsevier Ltd",

}

RIS

TY - JOUR

T1 - A penalised piecewise-linear model for non-stationary extreme value analysis of peaks over threshold

AU - Barlow, Anna Maria

AU - Mackay, Ed

AU - Eastoe, Emma

AU - Jonathan, Philip

PY - 2023/1/1

Y1 - 2023/1/1

N2 - Metocean extremes often vary systematically with covariates such as direction and season. In this work, we present non-stationary models for the size and rate of occurrence of peaks over threshold of metocean variables with respect to one- or two-dimensional covariates. The variation of model parameters with covariate is described using a piecewise-linear function in one or two dimensions, defined with respect to pre-specified node locations on the covariate domain. Parameter roughness is regulated to provide optimal predictive performance, assessed using cross-validation, within a penalised likelihood framework for inference. Parameter uncertainty is quantified using bootstrap resampling. The models are used to estimate extremes of storm-peak significant wave height with respect to direction and season for a site in the northern North Sea. A covariate representation based on a triangulation of the direction-season domain with six nodes gives good predictive performance. The penalised piecewise-linear framework provides a flexible representation of covariate effects at reasonable computational cost.

AB - Metocean extremes often vary systematically with covariates such as direction and season. In this work, we present non-stationary models for the size and rate of occurrence of peaks over threshold of metocean variables with respect to one- or two-dimensional covariates. The variation of model parameters with covariate is described using a piecewise-linear function in one or two dimensions, defined with respect to pre-specified node locations on the covariate domain. Parameter roughness is regulated to provide optimal predictive performance, assessed using cross-validation, within a penalised likelihood framework for inference. Parameter uncertainty is quantified using bootstrap resampling. The models are used to estimate extremes of storm-peak significant wave height with respect to direction and season for a site in the northern North Sea. A covariate representation based on a triangulation of the direction-season domain with six nodes gives good predictive performance. The penalised piecewise-linear framework provides a flexible representation of covariate effects at reasonable computational cost.

KW - Extreme

KW - Non-stationary

KW - Covariate

KW - Significant wave height

KW - Penalised likelihood

U2 - 10.1016/j.oceaneng.2022.113265

DO - 10.1016/j.oceaneng.2022.113265

M3 - Journal article

VL - 267

JO - Ocean Engineering

JF - Ocean Engineering

SN - 0029-8018

M1 - 113265

ER -