摘要
利用极值理论给出了一种新的解决非寿险精算中巨额损失保费厘定问题的方法。在建模过程首先给出了极值理论的最大吸引域检验问题,然后利用不同方法讨论了最优门限值的选取问题,并在POT模型下利用广义帕累托分布对巨额损失分布进行拟合。然后在假设损失次数服从泊松分布的条件下,在复合泊松分布的框架下讨论了险位超赔再保险的纯保费计算问题。
In this paper, a new method of non-life actuary with extremely large loss has been presented by employing the extreme value theory. In the modeling process, first, the test of extreme value condition has been introduced, and then different methods of optimal choice of the threshold were given. Then we fit the extremely large loss data with a generalized pareto distribution. Under the hypothesis of compound Poisson distribution, a new method of calculation the pure premium of Execss-of-Loss insurance was given.
出处
《数理统计与管理》
CSSCI
北大核心
2010年第2期336-347,共12页
Journal of Applied Statistics and Management
关键词
极值理论
最大吸引域
广义帕累托分布
复合泊松分布
extreme value theory, the maximum attraction of GEV, generalized Pareto distribution, compound poisson distribution
