ABSTRACT
In Chapter 14 on “General Insurance Pricing,” Boucher & Charpentier discuss regression techniques suitable for pricing with cross-sectional data. A cross-sectional dataset observes each subject in the sample (for instance, a policy(holder)) only once. Each subject is therefore described by a single response, say Yi for subject i, and vector with covariate informa-
function of risk factors, within an appropriate distributional framework. For pricing in general insurance, the actuary builds such models for a (cross-sectional) dataset with claim counts on the one hand and claim severities on the other hand as response variables, obtained by following policyholders during a single time period. The result is a tariff based on risk classification through regression modeling. When the explanatory variables used as rating factors express a priori correctly measurable information about the policyholder (or, for instance, the vehicle or the insured building), the system is called an a priori classification scheme. The examples in Chapter 14 illustrate this idea and use, for instance, age of the driver and age of the car to explain the number of claims registered by a policyholder.
