摘要
泊松分布和负二项分布常用于拟合保险索赔次数.它们和二项分布统称为(a,b,0)分布族.本文对(a,b,0)分布族进行了研究,然后在此基础上给出了(a,b,0)分布离散型随机变量是服从泊松分布,还是服从负二项分布或二项分布的检验方法.本文基于我国某家保险公司的索赔次数数据进行了实证分析,并对检验的功效进行了模拟研究.
Poisson and negative binomial distributions are usually used to fit claim frequencies. The two discrete distributions and binomial distribution totally constitute a class of discrete non-negative integer-valued distributions called （a, b, 0） family. In this paper, we discuss （a, b, 0） family first. Then we advance a hypothesis testing method to distinguish which distribution it is from when given a set of claim frequency data. In section 3, we apply the method on real claim frequency data from some domestic insurer and make a simulative study on the power of the method.
出处
《应用概率统计》
CSCD
北大核心
2005年第3期315-321,共7页
Chinese Journal of Applied Probability and Statistics
基金
上海市科委基础研究重点项目(02DJ14063)国家社会科学基金项目(03BTJ006)资助.
