BAYESIAN ANALYSIS OF THE COMPOUND COLLECTIVE MODEL; THE VARIANCE PREMIUM PRINCIPLE WITH EXPONENTIAL POISSON AND GAMMA-GAMMA DISTRIBUTIONS

The distribution of the aggregate claim size is the considerable importance in insurance theory since, for example, it is needed as an input in premium calculation principles and reserve calculation which plays an important paper in ruin theory. In this paper a Bayesian study for the collective risk model by incorporating a prior distribution for both, the parameter of the claim number distribution and the parameter of the claim size distribution is made and applied to the variance premium principle. Later a sensitivity study is to carry out on both parameters using Bayesian global robustness. Despite the complicated form of the collective risk model it is shown how the robustness study can be treated in an easy way. We illustrate the results obtained with numerical examples.