摘要
本文应用我国一家保险公司的实际数据 ,对各种可以反映保单持有人索赔次数的模型 (包括负二项模型、泊松 -逆高斯模型、二元风险模型、三元风险模型、二项 -贝塔模型和负二项 -帆塔模型 )分别进行了拟合 ,结果表明三元风险模型拟合效果最好 。
This paper discusses the claim number models and their application in Chinese automobile third party liability insurance.The paper use Negative-Binomial model,Poisson Inverse-Gassian model,Two-Point model,Three-Point model,Binomial-Beta model,and Negative-Binomial model to fit the data from a Chinese insurance company.Theresult shows that Three-Point model fit the data best,so the optimal BMS is constructed using this model.
出处
《数理统计与管理》
CSSCI
北大核心
2001年第3期60-65,共6页
Journal of Applied Statistics and Management
基金
CCU IPP资助
NSFC资助
