Prediction of Information Leakage by Cyber Incidents Via Classical Compound Risk Models

We show that classical insurance models based on some compound distributions can well predict information leakage by cyber incidents with reducing the computational cost thanks to the model’s simplicity. We use the negative binomial distribution, a renewal process and a Hawkes process as the compound point processes. Based on our models, we calculate VaR and TVaR for the total breach sizes per period. We compared them with the actual breach size by back-testing to verify the accuracy of our method