Modeling and Management of Nonlinear Dependencies-Copulas in Dynamic Financial Analysis

We study the influence of nonlinear dependencies on a non-life insurer's risk and return profile. To achieve this, we integrate several copula models in a dynamic financial analysis framework and conduct numerical tests. We also test risk management strategies in response to adverse outcomes. Nonlinear dependencies have a crucial influence on the insurer's risk profile that can hardly be affected by the analyzed management strategies. We find large differences in risk assessment for the ruin probability and for the expected policyholder deficit. This has important implications for insurers, regulators, and rating agencies that use these measures as a foundation for internal risk models, capital standards, and ratings. Copyright (c) The Journal of Risk and Insurance, 2009.