Combining generalized linear models and credibility models in practice

Abstract

Rating of non-life insurance contracts commonly employs multiplicative models, which are estimated by generalized linear models (GLMs); another useful tool for rate making are credibility models. The object of this paper is to demonstrate how these can be combined in practice, to solve the problem with multi-level factors – rating factors with too many levels for GLM estimation. In particular, we consider car model classification in motor insurance, using data from a Swedish insurance company.