Utility of classical insurance risk models for measuring the risks of cyber incidents

Abstract

We demonstrate that the classical insurance risk models yield significant advantages in the context of cyber risk analysis. This model exhibits commendable attributes in terms of both computational efficiency and predictive capabilities. Utilizing several compound point risk models, we derive the conditional Value-at-Risk and Tail Value-at-Risk predictions for the cumulative breach size within specified time intervals. To verify the reliability of our method, we conduct backtesting exercises, comparing our predictions with actual breach sizes.