摘要
使用组合模型模拟单峰厚尾型保险损失数据是非常有效的方法.鉴于在非寿险合约中一般都具有免赔条款的特征,构建一类截断指数威布尔-帕累托组合模型,讨论模型的相关统计性质,然后利用R语言对仿真数据进行参数估计及模型检验.最后,使用丹麦火险数据进行分布拟合,实证结果表明,截断指数威布尔-帕累托组合模型具有更优的拟合效果.
It is very effective to simulate the insurance loss data with leptokurtic shape using the composite model. Note that most non-life insurance contracts generally have the deductibles, we construct the truncated composite exponentiated Weibull-Pareto model and study the related statistical properties. The parameter estimation and model checking of simulation data are given using the R language. Finally, we illustrate an application to the Danish fire loss data, it is shown that the newly proposed model gives a better fit.
出处
《数学的实践与认识》
北大核心
2017年第24期178-185,共8页
Mathematics in Practice and Theory
基金
教育部人文社会科学研究青年基金(15YJC910001)
关键词
截断指数威布尔-帕累托组合模型
最大似然估计
丹麦火险损失数据
the truncated composite exponentiated Weibull-Pareto model
maximum like-lihood estimation
Danish fire loss data