Title: Regression Tree Credibility Model
Authors: Liqun Diao - University of Waterloo (Canada)
Chengguo Weng - University of Waterloo (Canada) [presenting]
Abstract: This work applies learning techniques to credibility theory and proposes a regression tree based algorithm to integrate covariate information for credibility premium prediction. The algorithm recursively binary partitions a collective of individual risks into mutually exclusive sub-collectives, and consequently applies the classical Buhlmann-Straub credibility formula for the prediction of individual net premium within each sub-collective. It provides a flexible way to integrate covariate information into individual net premium prediction. It is appealing for capturing non-linear and/or interaction covariate effects. It automatically selects influential covariate variables for premium prediction with no additional ex-ante variable selection procedure required. The superiority in prediction accuracy of the proposed model is demonstrated by extensive simulation studies. An application to the U.S. Medicare data is presented.