Classification Ratemaking via Quantile Regression and a Comparison with Generalized Linear Models

In non-life insurance, it is important to develop a loaded premium for individual risks, as the sum of a pure premium (expected value of loss) and a safety loading or risk margin. In actuarial practice, this process is known as classification ratemaking and is performed usually via Generalized Linear Model. The latter permits an estimate of individual pure premium and safety loading both; however, the goodness of the estimates are strongly related to the compliance of the model assumption with the empirical distribution. In order to investigate the individual pure premium, we introduce an alternative pricing model based on Quantile Regression, to perform a working classification ratemaking with weaker assumptions and, then, more performing for risk margin evaluation.