Modeling nonstationary extremes of storm severity: Comparing parametric and semiparametric inference

This article compares the modeling of nonstationary extreme events using parametric models with local parametric and semiparametric approaches also motivated by extreme value theory. Specifically, three estimators are compared based on (a) (local) semiparametric moment estimation, (b) (local) maximum likelihood estimation, and (c) spline-based maximum likelihood estimation. Inference is performed in a sequential manner, highlighting the synergies between the different approaches to estimating extreme quantiles, including the T-year level and right endpoint when finite. We present a novel heuristic to estimate nonstationary extreme value threshold with exceedances varying on a circular domain, and hypothesis-testing procedures for identifying max-domain of attraction in the nonstationary setting. Bootstrapping is used to estimate nonstationary confidence bounds throughout. We provide step-by-step guides for estimation, and explore the different inference strategies in application to directional modeling of hindcast storm peak significant wave heights recorded in the North Sea. © 2021 The Authors. Environmetrics published by John Wiley & Sons, Ltd.

[1]  P. Jonathan,et al.  Threshold modeling of nonstationary extremes , 2016 .

[2]  P. Jonathan,et al.  Bayesian inference for non-stationary marginal extremes , 2016 .

[3]  Richard L. Smith.  Estimating tails of probability distributions , 1987 .

[4]  Bayesian P-spline mixture modeling of extreme forest temperatures , 2012 .

[5]  D. Randell,et al.  Flexible covariate representations for extremes , 2020, Environmetrics.

[6]  Cláudia Neves,et al.  Reiss and Thomas' automatic selection of the number of extremes , 2004, Comput. Stat. Data Anal..

[7]  Jun Yan,et al.  Extreme Value Modeling and Risk Analysis : Methods and Applications , 2015 .

[8]  Richard L. Smith,et al.  Models for exceedances over high thresholds , 1990 .

[9]  Hilde Haakenstad,et al.  A high‐resolution hindcast of wind and waves for the North Sea, the Norwegian Sea, and the Barents Sea , 2011 .

[10]  P. Jonathan,et al.  Estimating surge in extreme North Sea storms , 2018 .

[11]  António B. Pereira,et al.  Detecting finiteness in the right endpoint of light-tailed distributions , 2010 .

[12]  L. Haan,et al.  A moment estimator for the index of an extreme-value distribution , 1989 .

[13]  Eric P. Smith,et al.  An Introduction to Statistical Modeling of Extreme Values , 2002, Technometrics.

[14]  Alan Y. Chiang,et al.  Generalized Additive Models: An Introduction With R , 2007, Technometrics.

[15]  R. Fisher,et al.  Limiting forms of the frequency distribution of the largest or smallest member of a sample , 1928, Mathematical Proceedings of the Cambridge Philosophical Society.

[16]  George Z. Forristall,et al.  On the Use of Directional Wave Criteria , 2004 .

[17]  Malcolm R Leadbetter,et al.  On a basis for 'Peaks over Threshold' modeling , 1991 .

[18]  L.F.M. deHaan.  On regular variation and its application to the weak convergence of sample extremes , 1970 .

[19]  L. Haan,et al.  Residual Life Time at Great Age , 1974 .

[20]  C. Neves,et al.  A general estimator for the right endpoint with an application to supercentenarian women’s records , 2014, 1412.3972.

[21]  L. Haan,et al.  On tail trend detection: modeling relative risk , 2011, 1106.4149.

[22]  AbuBakr S. Bahaj,et al.  On the use of discrete seasonal and directional models for the estimation of extreme wave conditions , 2010 .

[23]  Philip Jonathan,et al.  Threshold modelling of spatially dependent non‐stationary extremes with application to hurricane‐induced wave heights , 2011 .

[24]  Stuart Coles,et al.  Directional Modelling of Extreme Wind Speeds , 1994 .

[25]  Melisa Menéndez,et al.  Seasonality and duration in extreme value distributions of significant wave height , 2008 .

[26]  Jonathan A. Tawn,et al.  Statistics for Extreme Sea Currents , 1997 .

[27]  B. Gnedenko.  Sur La Distribution Limite Du Terme Maximum D'Une Serie Aleatoire , 1943 .

[28]  Philip Jonathan,et al.  Bayesian inference for nonstationary marginal extremes , 2016 .

[29]  J. Pickands.  Statistical Inference Using Extreme Order Statistics , 1975 .

[30]  M. D. Ugarte,et al.  Projections of cancer mortality risks using spatio-temporal P-spline models , 2012, Statistical methods in medical research.

[31]  Miguel A. Losada,et al.  The selection of directional sectors for the analysis of extreme wind speed , 2019 .

[32]  Philip Jonathan,et al.  The effect of directionality on extreme wave design criteria , 2007 .

[33]  Philip Jonathan,et al.  The Effect of Directionality on Northern North Sea Extreme Wave Design Criteria , 2008 .

[34]  P. Jonathan,et al.  Statistics of extreme ocean environments: Non-stationary inference for directionality and other covariate effects , 2016, 1807.10542.

[35]  Anthony C. Davison,et al.  Statistics of Extremes , 2015, International Encyclopedia of Statistical Science.

[36]  Philip Jonathan,et al.  On the estimation and application of directional design criteria , 2019 .

[37]  Isabel Fraga Alves,et al.  ESTIMATION OF THE FINITE RIGHT ENDPOINT IN THE GUMBEL DOMAIN , 2013, 1306.1452.

[38]  A. Davison,et al.  Generalized additive modelling of sample extremes , 2005 .

[39]  Kees Mulder,et al.  Circular Statistics in R , 2015 .

[40]  M. Gomes,et al.  Asymptotic comparison of the mixed moment and classical extreme value index estimators , 2008 .