Generalised Additive Modelling of Auto Insurance Data with Territory Design: A Rate Regulation Perspective

Pricing using a Generalised Linear Model is the gold standard in auto insurance industry and rate regulation. Additive applications pricing are receiving increasing attention from academic researchers actuarial professionals. The practice has constantly shown evidence of significantly different premium rates among rating territories. In this work, we build predictive models for claim frequency severity synthetic Usage Based Insurance (UBI) dataset variables. First, conduct territorial clustering based on each location’s counts amounts by grouping those locations into smaller set, defined as cluster purposes. After clustering, incorporate these clusters our model to determine risk relativity factor level. Through modelling, have successfully identified key factors that may be helpful regulation UBI. Our work aims fill gap between individual-level UBI database provides insights consistency traditional variables pricing. main contribution outline how GAM can address more complicated functionality interactions them. We also contribute demonstrating territory problem construct territories find high annual mileage driven almost three times associated with low level, which implies its importance calculation. Overall, provide regulated through factors, additional datasets derived basic units driver’s location.