Tail-Aware Distribution Modeling for Extreme Weather Losses: Generalized Pareto Analysis of NOAA Data
Main Article Content
Abstract
Extreme events and rare outcomes are difficult to model because they occur infrequently, show strong skewness, and often produce consequences much larger than ordinary observations. This study develops and tests an empirical probability-distribution fitting framework for modeling extreme weather losses and rare outcomes using the National Oceanic and Atmospheric Administration (NOAA) Storm Events dataset. The analysis uses 1,962,332 storm-event records from 1950 to 2025 and focuses on 450,823 positive economic-loss observations after converting reported property and crop damage values from Thousands/Millions/Billions (K/M/B)-coded strings into numeric United States (U.S.) dollars. Candidate distributions include normal, lognormal, exponential, Weibull, Pareto, generalized extreme value, and generalized Pareto models. Parameters are estimated by maximum likelihood estimation (MLE), while performance is evaluated using log-likelihood, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Kolmogorov–Smirnov (KS) distance, graphical diagnostics, threshold sensitivity, and high-quantile tail-risk estimates. The empirical results show strong heavy-tailed behavior: positive damage has a median of $10.00 thousand, a 99th percentile of $10.59 million, a 99.9th percentile of $200.00 million, and a maximum reported value of $17.90 billion. The lognormal distribution provides the strongest full-body fit among conventional models, but it substantially underestimates the far upper tail. By contrast, the generalized Pareto peaks-over-threshold model at the 95th percentile produces tail estimates close to the empirical 99th and 99.9th percentiles, with a Kolmogorov–Smirnov (KS) distance of 0.076 for exceedances. The study contributes a transparent, reproducible framework for fitting probability distributions to rare event losses and shows that separate body and tail modeling is necessary for reliable risk estimation.
Article Details
Issue
Section

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
How to Cite
References
[1]. AghaKouchak, A., Easterling, D., Hsu, K., Schubert, S., & Sorooshian, S. (Eds.). (2013). Extremes in a changing climate: Detection, analysis and uncertainty. Springer.
[2]. Aminzadeh, M. S., & Deng, M. (2017). Bayesian predictive modeling for exponential-Pareto composite distribution. Variance, 12(1), 59-78.
[3]. Bader, B., Yan, J., & Zhang, X. (2018). Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate. The Annals of Applied Statistics, 12(1), 310-329.
[4]. Beirlant, J., Fraga Alves, I., Gomes, I., & Meerschaert, M. M. (2016). Extreme value statistics for truncated Pareto-type distributions. Extremes, 19, 1-29.
[5]. Charpentier, A., & Flachaire, E. (2021). Pareto models for risk management. In G. Dufrenot and T. Matsuki (Eds.), Recent econometric techniques for macroeconomic and financial data (pp. 125-145). Springer.
[6]. Chavez-Demoulin, V., & Davison, A. C. (2012). Modeling time series extremes. REVSTAT - Statistical Journal, 10(1), 109-133.
[7]. Davison, A. C., & Huser, R. (2015). Statistics of extremes. Annual Review of Statistics and Its Application, 2, 203-235.
[8]. Davison, A. C., Huser, R., & Thibaud, E. (2019). Spatial extremes. In A. E. Gelfand, M. Fuentes, J. A. Hoeting, & R. L. Smith (Eds.), Handbook of environmental and ecological statistics (pp. 711-744). CRC Press.
[9]. Davison, A. C., Padoan, S. A., & Ribatet, M. (2012). Statistical modeling of spatial extremes. Statistical Science, 27(2), 161-186.
[10]. Deng, M. (2021). Bayesian predictive analysis of natural disaster losses. Risks, 9(1), 12.
[11]. Gilleland, E., & Katz, R. W. (2016). extRemes 2.0: An extreme value analysis package in R. Journal of Statistical Software, 72(8), 1-39.
[12]. He, Y. I., Chen, R., Lee, T. T., & Lou, K. (2022). Risk analysis via generalized Pareto distributions. Risks, 10(6), 117.
[13]. Huser, R., & Wadsworth, J. L. (2022). Advances in statistical modeling of spatial extremes. WIREs Computational Statistics, 14(1), e1537.
[14]. Katz, R. W. (2010). Statistics of extremes in climate change. Climatic Change, 100(1), 71-76.
[15]. Kousky, C. (2014). Informing climate adaptation: A review of the economic costs of natural disasters. Energy Economics, 46, 576-592.
[16]. Merz, B., Aerts, J. C. J. H., Arnbjerg-Nielsen, K., Baldi, M., Becker, A., Bichet, A., Bloschl, G., Bouwer, L. M., Brauer, A., Cioffi, F., Delgado, J. M., Gocht, M., Guzzetti, F., Harrigan, S., Hirschboeck, K., Kilsby, C., Kron, W., Kwon, H. H., Lall, U., Merz, R., Nissen, K., Salvatti, P., Swierczynski, T., Ulbrich, U., Viglione, A., Ward, P. J., Weiler, M., Wilhelm, B., & Nied, M. (2014). Floods and climate: Emerging perspectives for flood risk assessment and management. Natural Hazards and Earth System Sciences, 14, 1921-1942.
[17]. Naveau, P., Huser, R., Ribereau, P., & Hannart, A. (2016). Modeling jointly low, moderate, and heavy rainfall intensities without a threshold selection. Water Resources Research, 52(4), 2753-2769.
[18]. NOAA National Centers for Environmental Information. (2022). Storm Events Database. National Oceanic and Atmospheric Administration.
[19]. Northrop, P. J., & Coleman, C. L. (2014). Improved threshold diagnostic plots for extreme value analyses. Extremes, 17, 289-303.
[20]. Opitz, T. (2016). Modeling asymptotically independent spatial extremes based on Laplace random fields. Spatial Statistics, 16, 1-18.
[21]. Paprotny, D., Sebastian, A., Morales-Napoles, O., & Jonkman, S. N. (2018). HANZE: A pan-European database of exposure to natural hazards and damaging historical floods since 1870. Earth System Science Data, 10, 565-581.
[22]. Papastathopoulos, I., & Tawn, J. A. (2013). Extended generalized Pareto models for tail estimation. Journal of Statistical Planning and Inference, 143(1), 131-143.
[23]. Rootzen, H., & Katz, R. W. (2013). Design life level: Quantifying risk in a changing climate. Water Resources Research, 49(9), 5964-5972.
[24]. Scarrott, C., & MacDonald, A. (2012). A review of extreme value threshold estimation and uncertainty quantification. REVSTAT - Statistical Journal, 10(1), 33-60.
[25]. Schneider, L. F., Krajina, A., & Krivobokova, T. (2021). Threshold selection in univariate extreme value analysis. Extremes, 24, 881-913.
[26]. Seneviratne, S. I., Zhang, X., Adnan, M., Badi, W., Dereczynski, C., Di Luca, A., Ghosh, S., Iskandar, I., Kossin, J., Lewis, S., Otto, F., Pinto, I., Satoh, M., Vicente-Serrano, S. M., Wehner, M., & Zhou, B. (2021). Weather and climate extreme events in a changing climate. In V. Masson-Delmotte et al. (Eds.), Climate change 2021: The physical science basis. Cambridge University Press.
[27]. Singirankabo, E., & Iyamuremye, A. (2022). Modeling extreme rainfall events in Kigali city using generalized Pareto distribution. Meteorological Applications, 29(3), e2076.
[28]. Wirtz, A., Kron, W., Loewe, P., & Steuer, M. (2014). The need for data: Natural disasters and the challenges of database management. Natural Hazards, 70, 135-157.
[29]. Zhang, M., Wen, R., Ren, Y., & Ji, K. (2021). Application of generalized Pareto distribution for modeling extreme ground motion in probabilistic seismic hazard analysis. Natural Hazards, 107, 1527-1547.
[30]. Zscheischler, J., Martius, O., Westra, S., Bevacqua, E., Raymond, C., Horton, R. M., van den Hurk, B., AghaKouchak, A., Jezequel, A., Mahecha, M. D., Maraun, D., Ramos, A. M., Ridder, N. N., Thiery, W., & Vignotto, E. (2020). A typology of compound weather and climate events. Nature Reviews Earth & Environment, 1, 333-347.