Lluís Bermúdez, Michel Denuit, and Jan Dhaene
Abstract
This paper examines an integrated ratemaking scheme including {a priori} risk classification and {a posteriori} experience rating. In order to avoid the high penalties implied by the quadratic loss function, the symmetry between the overcharges and the undercharges is broken by introducing parametric loss functions of exponential type.
Key words and phrases: quadratic loss function, exponential loss function, credibility estimation, explanatory variables, experience rating, risk classification.
Corresponding Author:
Michel Denuit
Institut de Statistique,
Universite Catholique de Louvain,
Voie du Roman Pays 20,
B-1348 Louvain-la-Neuve
© Copyright Absalom Press, Inc.