Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
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 -