Something Old, Something New in Classification Ratemaking With a Novel Use of GLMs for Credit Insurance

This paper discusses some methods that can be used to calculate classification relativities and reduce the error that would otherwise occur by using one-way analysis. Section 2 will discuss the problem of risk classification analysis from a mathematical and statistical viewpoint and show some of the implied solutions from these approaches. This exposition revisits the work pioneered in the USA by Bailey, Bailey and Simon, and Brown, which are the foundations of American casualty practice in the area of classification ratemaking. We will then revisit another technique based on Generalized Linear Modeling (GLM) in Section 3 and discuss the advantages of implementing this technique. For those who have a strong background in classification ratemaking and GLM, we recommend skipping to Sections 4 and 5, where we present an application of this technique to credit insurance and discuss the results.