吉林大学学报(理学版)

• 数学 • 上一篇    下一篇

相依计数变量风险模型的大偏差

宇世航1,2, 王德辉2, 魏蕴波1   

  1. 1. 齐齐哈尔大学 理学院, 黑龙江 齐齐哈尔 161006; 2. 吉林大学 数学学院, 长春 130012  
  • 收稿日期:2012-11-21 出版日期:2013-07-26 发布日期:2013-08-06
  • 通讯作者: 宇世航 E-mail:qqhrysh@163.com

Large Deviations in Risk Models Based onDependent

YU Shihang1,2, WANG Dehui2, WEI Yun bo1   

  1. 1. College of Science, Qiqihar University, Qiqihar 161006, Heilongjiang Province, China;2. College of Mathematics, Jilin University, Changchun 130012, China
  • Received:2012-11-21 Online:2013-07-26 Published:2013-08-06
  • Contact: YU Shihang E-mail:qqhrysh@163.com

摘要:

考虑每期索赔计数变量之间基于泊松AR(1)相依结构的离散风险模型, 利用特征函数的唯一性, 得到了其累积索赔总额的概率分布等价形式, 并建立了重尾索赔下索赔总额的精细大偏差.

Abstract:

We considered discretetime risk model in which a dependent structure of Poisson AR(1) is introduced between the claim numbers for each period. With the help of uniqueness of characteristic function, we derived probability distribution equivalent form of accumulated aggregate claim amount, and then established large deviation results for the sum of heavytailed claim sizes.

Key words: discretetime risk models, Poisson AR(1) process, heavytailed distribution, large deviation

中图分类号: 

  • O212.7