Geographical Ratings with Spatial Random Effects in a Two-Part Model

Abstract

Rating areas are commonly used to capture unexplained geographical variability of claims in insurance pricing. A new method for defining rating areas is proposed using a two-part generalized geoadditive model that models spatial effects smoothly using Gaussian Markov random fields. The first part handles zero/nonzero expenses in a logistic model; the second handles nonzero expenses (on log-scale) in a linear model. Both models are fit with R package INLA for Bayesian infer-ences. The resulting spatial effects are used to construct more representative ratings. The methodology is illustrated with simulated data based on zipcode areas, but modeled on zipcode- or county-level.

Volume
13
Issue
1
Page
141-160
Year
2020
Keywords
Geoadditive model; INLA; rating area; territory analysis
Categories
Financial and Statistical Methods
Statistical Models and Methods
Bayesian Methods
Publications
Variance
Authors
Elizabeth D. Schifano
Chun Wang
Jun Yan

Keep up with the latest CAS news