🤖 AI Summary
This paper addresses the inadequate modeling of multiregional loss dependence structures in cross-regional catastrophe (CAT) bond pricing. We develop a unified multi-regional pricing framework that systematically characterizes independent, proportional, and general bivariate extreme-value dependence structures, and integrates the Wang transform to explicitly incorporate market risk preferences. Using historical PCS loss data, we empirically assess how alternative dependence assumptions affect CAT bond prices and derive a closed-form normal approximation solution. Results demonstrate that the choice of dependence structure significantly impacts pricing outcomes, while the normal approximation maintains high accuracy under real-world data. This study provides a theoretically consistent, computationally efficient, and operationally practical valuation tool for the design and risk management of multi-regional CAT bonds.
📝 Abstract
The insurance-linked securities (ILS) market, as a form of alternative risk transfer, has been at the forefront of innovative risk-transfer solutions. The catastrophe bond (CAT bond) market now represents almost half of the entire ILS market and is growing steadily. Since CAT bonds are often tied to risks in different regions, we follow this idea by constructing different pricing models that incorporate various scenarios of dependence between catastrophe losses in different areas. Namely, we consider independent, proportional, and arbitrary two-dimensional distribution cases. We also derive a normal approximation of the prices. Finally, to include the market price of risk, we apply Wang's transform. We illustrate the differences between the scenarios and the performance of the approximation on the Property Claim Services data.