4 résultats pour « robustness »

Distributionally robust insurance under the Wasserstein distance

This paper explores optimal insurance contracting for a decision maker facing ambiguous loss distributions. Using a p-Wasserstein ball around a benchmark distribution and a convex distortion risk measure, the indemnity function and worst-case distribution are derived. Numerical examples highlight the sensitivity of worst-case distributions to model parameters.Distributionally robust insurance under the Wasserstein distance

Worst‑Case Reinsurance Strategy with Likelihood Ratio Uncertainty

The study delves into optimizing reinsurance amidst uncertainty, aiming to minimize insurer's worst-case loss. It establishes a connection between optimal strategies under a reference measure and those in worst-case scenarios, applicable to tail risk quantification. Conditions for common optimal solutions are provided, with applications to expectile risk measures explored. Cooperative and non-cooperative models are compared.

Risk Measures: Robustness, Elicitability, and Backtesting

"...we argue that... the median shortfall—that is, the median of the tail loss distribution—is a better option than the expected shortfall for setting the Basel Accords capital requirements due to statistical and economic considerations such as capturing tail risk, robustness, elicitability, backtesting, and surplus invariance."