用指数复合逆伽马Pareto模型分析保险数据

摘要

指数模型已广泛应用于各种类型的数据建模,如生存数据以及保险索赔数据。然而,指数复合分布模型尚未被广泛探索。本文通过对与单参数逆伽玛Pareto复合分布函数相关的随机变量求指数,创造了一种改进的双参数逆伽玛Pareto复合模型,即指数复合逆伽马Pareto模型,并且很好地分析了挪威火险数据案例。得出双参数指数逆伽马Pareto模型在所有数据集的拟合优度测度方面优于单参数逆伽马Pareto模型的结论。Exponential models have been widely used to model various types of data, such as survival data and insurance claim data. However, exponential composite distribution models have not been widely explored. In this paper, an improved two-parameter inverse-gamma Pareto composite model, namely exponential composite inverse-gamma Pareto model, is created by taking exponents of random variables related to the single-parameter inverse-gamma Pareto composite distribution function, and a good case of Norwegian fire data is analyzed. It is concluded that the two-parameter exponential gamma Pareto model is superior to the single-parameter inverse-gamma Pareto model in the goodness of fit measure of all data sets.