基于尺度混合偏正态分布的稳健未决赔款准备金评估方法  

Robust Outstanding Claims Reserve Assessment Methods Based on Scale Mixtures of Skew-normal Distribution

作　　者：王明高[1] 孟生旺[2] WANG Ming-gao;MENG Sheng-wang(School of Statistics,Shandong Institute of Business and Technology,Yantai 264005,China;Center for Applied Statistics,Renmin University of China,Beijing 100872,China)

机构地区：[1]山东工商学院统计学院,山东烟台264005 [2]中国人民大学应用统计科学研究中心,北京100872

出　　处：《数理统计与管理》2021年第4期634-642,共9页Journal of Applied Statistics and Management

基　　金：国家社科基金项目(17BTJ031);教育部人文社会科学重点研究基地重大项目(16JJD910001);国家社科基金重大项目(16ZDA052)。

摘　　要：赔款准备金评估作为保险精算的一个重要研究领域,对于含有异常值的赔款数据,很多学者不断探索更加合理的赔款准备金评估方法,本文应用稳健统计方法,即假设赔款准备金模型中误差分布服从尺度混合偏正态分布构成的厚尾分布,降低异常值的影响,得到更加合理的评估结果。通过实际数据分析,并与已有的稳健赔款准备金统计方法进行比较,基于尺度混合偏正态分布的稳健赔款准备金评估方法取得比较好的评估结果,具有实际应用价值。Loss reserve assessment method is an important research field of insurance and actuarial science,many scholars continue to explore more reasonable method to deal with data containing abnormal claims.In this paper,we use the robust statistical method,that is,to assume that the error distribution in the loss reserve model has thick tail distribution of scale mixtures of skew-normal distribution,reduce the impact of abnormal values,to obtain more reasonable assessment results.Through the actual data analysis,and compared with the existing statistical methods,this paper has made a good evaluation result and has practical application value.

关 键 词：稳健统计 尺度混合偏正态分布 贝叶斯方法 

分 类 号：O212[理学—概率论与数理统计]