🤖 AI Summary
This study addresses the challenge of predicting economic losses from Atlantic tropical cyclones. We propose a multi-scale Bayesian hierarchical model that jointly forecasts seasonal total loss counts, individual hurricane occurrence probabilities, and associated monetary losses. Our method introduces a novel dual-temporal-scale modeling framework—integrating seasonal-frequency and event-level loss components—while incorporating climate indices (e.g., AMO, ENSO) into extreme-value statistics and multivariate joint distribution modeling. This design significantly enhances detection and quantification of rare, high-impact events. Through robust cross-validation and comprehensive benchmarking against state-of-the-art models, our approach achieves superior performance across multiple evaluation metrics. Beyond improved point prediction accuracy, it delivers interpretable and well-calibrated probabilistic uncertainty estimates. The resulting framework provides a reliable, decision-ready tool for catastrophe risk management and climate adaptation planning.
📝 Abstract
Bayesian hierarchical models are proposed for modeling tropical cyclone characteristics and their damage potential in the Atlantic basin. We model the joint probability distribution of tropical cyclone characteristics and their damage potential at two different temporal scales, while taking several climate indices into account. First, a predictive model for an entire season is developed that forecasts the number of cyclone events that will take place, the probability of each cyclone causing some amount of damage, and the monetized value of damages. Then, specific characteristics of individual cyclones are considered to predict the monetized value of the damage it will cause. Robustness studies are conducted and excellent prediction power is demonstrated across different data science models and evaluation techniques.
