Modèle bayésien de tarification de l’assurance des flottes de véhicules

We are proposing a parametric model to price insurance for road vehicles belonging to a fleet. The tables of premiums presented in the article take into account past vehicle accidents, observable characteristics of the vehicles and fleets, and violations of the road safety code committed by drivers and transporters. The premiums are also adjusted according to accidents accumulated by the fleets over time. The model proposed accounts directly for explicit changes in the various components of the probability of accidents. It represents an extension of insurance models for individual premiums (Lemaire, 1985; Dionne and Vannase, 1989 and 1992; Pinquet, 1997 and 1998; Frangos and Vrontos, 2001; Purcaru and Denuit, 2003). The extension adds a “fleet” effect to the “vehicle” effect so as to measure the impact that the non-observable characteristics or actions of transporters have on truck accident rates. This form of pricing offers several advantages. It allows us to visualize what impact the behaviours of owners and drivers can have on the predicted rate of accidents and, consequently, on premiums. It measures the influence of traffic violations and accumulated accidents on insurance premiums from a different angle. Indeed, the effects of violations are obtained by means of the regression component, whereas the effects of accidents are derived from unexplained residuals of the regression on truck accidents via a Bayes pricing model.