罚金函数和值函数及其应用：红利再保策略-概率论与数理统计专业论文.docx

I I 摘 要 本文讨论两个保险风险模型. 文章的第一部分讨论具有红利门槛的一个Sparre Andersen风险模型.从Gerber- Shiu贴现罚金函数的定义出发,利用数学分析中的一些基本思想方法获得了该模型 下的Gerber-Shiu贴现罚金函数的积分-微分方程及更新方程. 文章的第二部分讨论带有双门槛策略的一个离散时间模型,获得了Gerber-Shiu期 望贴现罚金函数和值函数的计算方法. 着重介绍了离散风险模型下一种新的再保 险策略:红利再保策略.即当保险公司的盈余变成负数时并且赤字不超过一定量时, 再保险公司必须支付保险公司一定数量的资金以解一时困境.反之,保险公司需要以 一部分红利作为再保费支付给再保险公司.应用期望贴现罚金函数和值函数,我们获 得了再保费的计算公式和破产前分给股东的红利期望现值的计算方法. 我们对保 险公司分配的红利期望现值和破产时间分别与无再保时对应的量进行了对比,并且 证明:在一定的条件下，红利再保策略不但能够显著提高破产前红利期望现值,而且 能够延长保险公司的经营寿命.文章运用了压缩映射及不动点原理对其中的数值问 题进行了计算,数值计算验证了我们的理论结果.这种新的再保险策略值得现实推 广. 关键词:Sparre Andersen风险模型; 离散时间风险模型; 积分-微分方程; 更 新方程; 红利再保策略; 值函数; Gerber-Shiu期望贴现罚金函数; 压缩映射. I II Abstract In this paper, we consider two risk models. In the first part of the paper, we discuss a Sparre Andersen risk model with a dividend threshold strategy. According to the definition of the Gerber-Shiu ex- pected discounted penalty function, we apply the methods of the mathematical analysis to derive the integro-differential equation and renewal equation of the Gerber-Shiu expected discounted penalty function. In the second part of the paper, we discuss the discrete time risk model with double thresholds strategy and offer the methods for computing the Gerber-Shiu expected discounted penalty function and value function. On this basis, we mainly propose a new type of reinsurance that we call dividend-reinsurance strategy. It is introduced as follows: the reinsurer is required to pay to the insurer a given amount of money whenever the surplus becomes negative and the deficit does not exceed the same amount of money, while in return the insurer pays to the reinsurer a fraction of all the dividends as reinsurance premium. Applying the expected discounted penalty function and value function, we obtain the calculation formula of reinsurance premium and the expected present value of dividends that are paid to the shareholders before ruin. We contrast the expected present value of dividends distributed by the company and ruin time in the presence of dividend