Numerical Study on The Optimal Reinsurance Problem with Two Correlated Risks

Abstract

This dissertation shows the work of the optimal reinsurance strategy with two correlated risks. By using R, this dissertation starts with generate random variables via Gaussian copula with different values of the correlation and the covariance matrix. After that, the model has been designed and the optimisation problem has been solved also by using R software. We have applied the study on three cases; where in the first case the two risks follow different distribution (exponential and Pareto distributions) and the second and third case the risks are following the same distribution either exponential distribution or Pareto distribution. We ended up with that: there is no specific strategy that is recommended for this situation and there is a relation between the randomness and the values of the correlation. The last thing is that the results should always be shown with respect to the two random variables and the ceded function.