TY - JOUR

T1 - Extreme value modelling of hurricane wind speeds.

AU - Coles, Stuart G.

AU - Casson, Edward

PY - 1998

Y1 - 1998

N2 - Estimating return levels of extreme wind speeds due to hurricanes presents both practical and analytical difficulties. The practical difficulty of collecting data has been resolved in the past by modelling simulated data--we adopt such an approach in this paper also. The analytical difficulties concern the problem of estimating the probabilities of events which are more extreme than those simulated. We follow common practice here also, using standard extreme value models to describe extreme tail behaviour. We differ from previous analyses of hurricane data in two respects. First, we use a model parameterisation which enables models fitted at different thresholds or at different sites to be easily compared. Second, we use maximum likelihood as the method of inference. This is found to produce results similar to those of previous studies, but enables the development of a spatial analysis which exploits similarities in the behaviour of the data from one site to another in order to improve the precision of estimation, and facilitates prediction at coastline locations other than those with simulated data.

AB - Estimating return levels of extreme wind speeds due to hurricanes presents both practical and analytical difficulties. The practical difficulty of collecting data has been resolved in the past by modelling simulated data--we adopt such an approach in this paper also. The analytical difficulties concern the problem of estimating the probabilities of events which are more extreme than those simulated. We follow common practice here also, using standard extreme value models to describe extreme tail behaviour. We differ from previous analyses of hurricane data in two respects. First, we use a model parameterisation which enables models fitted at different thresholds or at different sites to be easily compared. Second, we use maximum likelihood as the method of inference. This is found to produce results similar to those of previous studies, but enables the development of a spatial analysis which exploits similarities in the behaviour of the data from one site to another in order to improve the precision of estimation, and facilitates prediction at coastline locations other than those with simulated data.

KW - Extreme value theory

KW - Generalised Extreme Value (GEV) distribution

KW - Generalised Pareto distribution (GPD)

KW - Hurricanes

KW - Maximum Likelihood

KW - Spatial models

KW - Wind speeds

U2 - 10.1016/S0167-4730(98)00015-0

DO - 10.1016/S0167-4730(98)00015-0

M3 - Journal article

VL - 20

SP - 283

EP - 296

JO - Structural Safety

JF - Structural Safety

SN - 0167-4730

IS - 3

ER -