项目名称: 相依重尾随机变量和的渐近性及其在更新风险模型中的应用
项目编号: No.11426139
项目类型: 专项基金项目
立项/批准年度: 2015
项目学科: 数理科学和化学
项目作者: 于长俊
作者单位: 南通大学
项目金额: 3万元
中文摘要: 本项目将提出一些新的相依结构,并在这些相依结构下研究重尾随机变量和的渐近性。这类相依结构适正地包含了许多已有的相依结构,并且适合用来处理与随机变量之和相关的问题。我们将在这类结构下,提出一些研究随机变量部分和的尾概率的新方法,从而得到一般重尾随机变量和的渐近性和一般次指数随机变量和的渐近性。在这些结果中,前者将突破经典研究中对于独立性的限制,后者可以将以往相依结构下甚难研究的轻度重尾随机变量包含进来,因此这些结果具有重要的理论意义。 进一步地,我们会将上述结果应用到非标准的更新风险模型中,得到有限时破产概率和最终破产概率的渐近估计。相应地,我们的结果也包含了索赔额服从轻度重尾分布的情形。由于在很多情形下,索赔额均服从对数正态分布等轻度重尾分布,因此这些研究结果具有重要的应用价值。
中文关键词: 大额索赔;重尾分布;相依结构;更新风险模型;破产概率
英文摘要: This project will put forward some new dependence structures and will inveatigate the asymptotics of the sums of heavy-tailed random variables under these dependence structures. Dependence structures of this type properly contain many existing dependence structures, and are suitable in dealing with problems pertaining to the sums of the random variables.Under these types of structures,we will put forward some new methods to investigate the tail probability of the partial sums of the random variables, thus we will derive the asymptotics of the sums of general heavy-tailed random variables and general subexponential random variables. Among these results,the former part can breakthrough the restriction on independence in classical research,and the latter part will include the mild heavy-tailed random variables,therefore these results have great significance in theory . Further, we will apply these results into a non-standard renewal risk model,and get the asymptotic estimation of the finite-time ruin probability and the ultimate ruin probability. Correspondingly,our results also contains the case that the claim sizes follow mild heavy-tailed distributions. Since in many cases,the claim sizes follow mild heavy-taild distributions such as Lognormal distributions,these research results have important values
英文关键词: large claim amounts;heavy-tailed distributions;dependence structures;risk models;ruin probability