零调整分位数回归模型在车辆保险索赔额中的研究与应用

Research and Application of Zero-adjusted Quantile Regression Model for Estimating Claim Size in Vehicles Insures

车辆保险产品的定价一般会考虑保单持有人的索赔概率和期望索赔额等两个因素,零调整逆高斯回归模型作为解决这类问题的一个有力工具,由于变量分布的限定,从而具有一定的局限性.针对该问题,本文基于零调整逆高斯回归模型和分位数回归模型的思想,提出零调整分位数回归模型,并结合实际数据进行了拟合分析.与零调整逆高斯回归模型拟合的结果比较表明,零调整分位数回归模型可以作为研究车辆保险中索赔额的一个有力工具.

Claim probability and claim size are the factors for the proper pricing in vehicles insures.Though the zero-adjusted inverse Gaussian regression model is one of important statistical tools for the problem,it exists some drawbacks because of the choose about the distribution of the claim size variable.This paper,based on the ideas of the zero-adjusted inverse Gaussian regression model and the quantile regression model,first suggests the zero-adjusted quantile regression model and uses it to fit the actual motor insurance data.Through the analysis and comparing with the zero-adjusted inverse Gaussian regression model,the conclusion is that he zero-adjusted quantile regression model is appropriate for the claim size in vehicles insures.