Applying Graphical Models to Automobile Insurance Data
By Farrokh Guiahi
Analysis of insurance data provides input for making decisions regarding underwriting, pricing of insurance products, and claims, as well as profitability analysis. In this paper, we consider graphical modeling as a vehicle to reveal dependency structure of categorical variables used in the Australian Auto mobile data. The methodology developed here may supplement the traditional approach to ratemaking. Topics considered are the description of the automobile data set, preprocessing of the variables, visualization tools suitable for contingency tables, classical test of independence, loglinear models, the concept of conditional independence, and graphical modeling as a vehicle to explore the dependency structure among categorical variables, as well as a review of frequency rates by rating class.
Keywords: Categorical variables, visualization of categorical variables, mosaic plots, chi-square test, log-linear models, conditional independence, cliques, graphical modeling, frequency rates