Extreme value estimation using the likelihood-weighted method

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

Published

Standard

Extreme value estimation using the likelihood-weighted method. / Wada, R.; Waseda, T.; Jonathan, P.
In: Ocean Engineering, Vol. 124, 15.09.2016, p. 241-251.

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

Harvard

Wada, R, Waseda, T & Jonathan, P 2016, 'Extreme value estimation using the likelihood-weighted method', Ocean Engineering, vol. 124, pp. 241-251. https://doi.org/10.1016/j.oceaneng.2016.07.063

APA

Wada, R., Waseda, T., & Jonathan, P. (2016). Extreme value estimation using the likelihood-weighted method. Ocean Engineering, 124, 241-251. https://doi.org/10.1016/j.oceaneng.2016.07.063

Vancouver

Wada R, Waseda T, Jonathan P. Extreme value estimation using the likelihood-weighted method. Ocean Engineering. 2016 Sept 15;124:241-251. Epub 2016 Aug 6. doi: 10.1016/j.oceaneng.2016.07.063

Author

Wada, R. ; Waseda, T. ; Jonathan, P. / Extreme value estimation using the likelihood-weighted method. In: Ocean Engineering. 2016 ; Vol. 124. pp. 241-251.

Bibtex

@article{7105b8f5ae734f0382bfc97ed97b8d52,

title = "Extreme value estimation using the likelihood-weighted method",

abstract = "This paper proposes a practical approach to extreme value estimation for small samples of observations with truncated values, or high measurement uncertainty, facilitating reasonable estimation of epistemic uncertainty. The approach, called the likelihood-weighted method (LWM), involves Bayesian inference incorporating group likelihood for the generalised Pareto or generalised extreme value distributions and near-uniform prior distributions for parameters. Group likelihood (as opposed to standard likelihood) provides a straightforward mechanism to incorporate measurement error in inference, and adopting flat priors simplifies computation. The method's statistical and computational efficiency are validated by numerical experiment for small samples of size at most 10. Ocean wave applications reveal shortcomings of competitor methods, and advantages of estimating epistemic uncertainty within a Bayesian framework in particular. {\textcopyright} 2016 Elsevier Ltd",

keywords = "Bayes, Extreme, Group likelihood, Likelihood-weighted method, Uncertainty, Bayesian networks, Computational efficiency, Inference engines, Measurement errors, Numerical methods, Water waves, Weighted method, Uncertainty analysis, Bayesian analysis, estimation method, numerical method, ocean wave, uncertainty analysis",

author = "R. Wada and T. Waseda and P. Jonathan",

year = "2016",

month = sep,

day = "15",

doi = "10.1016/j.oceaneng.2016.07.063",

language = "English",

volume = "124",

pages = "241--251",

journal = "Ocean Engineering",

issn = "0029-8018",

publisher = "Elsevier Ltd",

}

RIS

TY - JOUR

T1 - Extreme value estimation using the likelihood-weighted method

AU - Wada, R.

AU - Waseda, T.

AU - Jonathan, P.

PY - 2016/9/15

Y1 - 2016/9/15

N2 - This paper proposes a practical approach to extreme value estimation for small samples of observations with truncated values, or high measurement uncertainty, facilitating reasonable estimation of epistemic uncertainty. The approach, called the likelihood-weighted method (LWM), involves Bayesian inference incorporating group likelihood for the generalised Pareto or generalised extreme value distributions and near-uniform prior distributions for parameters. Group likelihood (as opposed to standard likelihood) provides a straightforward mechanism to incorporate measurement error in inference, and adopting flat priors simplifies computation. The method's statistical and computational efficiency are validated by numerical experiment for small samples of size at most 10. Ocean wave applications reveal shortcomings of competitor methods, and advantages of estimating epistemic uncertainty within a Bayesian framework in particular. © 2016 Elsevier Ltd

AB - This paper proposes a practical approach to extreme value estimation for small samples of observations with truncated values, or high measurement uncertainty, facilitating reasonable estimation of epistemic uncertainty. The approach, called the likelihood-weighted method (LWM), involves Bayesian inference incorporating group likelihood for the generalised Pareto or generalised extreme value distributions and near-uniform prior distributions for parameters. Group likelihood (as opposed to standard likelihood) provides a straightforward mechanism to incorporate measurement error in inference, and adopting flat priors simplifies computation. The method's statistical and computational efficiency are validated by numerical experiment for small samples of size at most 10. Ocean wave applications reveal shortcomings of competitor methods, and advantages of estimating epistemic uncertainty within a Bayesian framework in particular. © 2016 Elsevier Ltd

KW - Bayes

KW - Extreme

KW - Group likelihood

KW - Likelihood-weighted method

KW - Uncertainty

KW - Bayesian networks

KW - Computational efficiency

KW - Inference engines

KW - Measurement errors

KW - Numerical methods

KW - Water waves

KW - Weighted method

KW - Uncertainty analysis

KW - Bayesian analysis

KW - estimation method

KW - numerical method

KW - ocean wave

KW - uncertainty analysis

U2 - 10.1016/j.oceaneng.2016.07.063

DO - 10.1016/j.oceaneng.2016.07.063

M3 - Journal article

VL - 124

SP - 241

EP - 251

JO - Ocean Engineering

JF - Ocean Engineering

SN - 0029-8018

ER -

Research

Associated organisational unit

Links

Text available via DOI:

Keywords