A fundamental goal of insurance companies is to calculate an appropriate insurance
price or premium corresponding to an insured in order to cover a certain risk. A well-known
method to calculate the premium is to multiply the conditional expectation of the claim
frequency with the expected cost of claims. Therefore, modelling frequency of claims, also
known in theory as count data, represents an essential step of non-life insurance pricing. As
sustained in Boucher and Guillen (2009), count regression analysis permits the identification
of the risk factors and the prediction of the expected frequency of claims given the risk
characteristics.