Title : The cost and severity of extreme natural disasters: What they mean for society and insurance
Abstract:
This research introduces a novel approach to estimating the economic impact of natural catastrophes (NatCats) by correlating log-loss severity, modeled using a Vasicek process, with the frequency of occurrences following a geometric Brownian motion. The novelty lies in combining these two processes to dynamically capture the relationship between the frequency and severity of catastrophic events, which provides a more comprehensive risk assessment. Unlike traditional models, this approach accounts for the joint dynamics of both losses and occurrences, offering a refined method for pricing tail events in NatCat insurance. The model's integration of these correlated processes enables more accurate pricing, particularly for extreme events, thus enhancing the ability of the insurance and reinsurance industries to assess and manage catastrophic risks. The model aligns risks with at-risk assets, helping policymakers prepare for and manage the financial challenges of natural disasters. Its parsimony, relying solely on occurrence and severity, allows for efficient and robust risk assessment, making it effective even with limited data.
