【作者】 张瑾；

【导师】 李秀芳；

【作者基本信息】 南开大学 ， 精算学， 2022， 硕士

【摘要】 国家最新统计数据显示,截至2021年底,我国65岁及以上老人占比14.2%,我国已提前进入“深度老龄化社会”,老年群体的健康保障成为当前社会面临的很大问题。在我国政府老年医疗及健康管理负担大、医疗费用支出负担增加的背景下,作为老年健康风险有力保障的商业健康保险发挥着日益重要的作用。如何开发更适合老年人的健康保险、解决老年人群的健康保障问题,是保险行业一直思考的问题和探索的方向。2013年,我国推出了首款老年防癌疾病保险,填补了我国老年健康保险市场的空白,提升了全民健康保障水平。疾病发生的概率及其损失分布是健康保险精算的定量基础,而目前国内关于疾病发生率测算和健康保险费率厘定的研究很少,相关的研究成果比较缺乏,定量基础的研究匮乏阻碍了健康保险精算技术的进一步发展。特别是对于老年防癌疾病保险来说,针对的人群是疾病高发的老年人,癌症的发病率高、波动性大、未来的发展趋势存在很大的不确定性,合理的基础数据和恰当的精算定价技术的重要性尤其高,直接关系到这类产品能否在我国健康保险市场中持续稳定的发展,因此,对于癌症发生率测量和老年防癌疾病保险定价问题的研究具有重要的理论意义和实践价值。本文在总结归纳国内外疾病发生率研究与疾病保险定价理论的基础之上,对老年人群癌症发生率的测量进行研究,建立了适合我国老年人群癌症发生率的动态随机模型,并提出了一种新的老年防癌疾病保险的定价方法,将时间变动带来的癌症发生率变化趋势的不确定性体现在这种定价方法中。本文的具体内容如下:(1)基于我国50-75岁人群癌症发生率数据的动静态特点,引入Lee-Carter模型等六种随机动态模型对癌症发生率及其未来变动进行测量,选择其中拟合效果最好的模型来度量老年人群的癌症疾病风险。(2)基于对曼联方法等三种传统精算定价方法的总结和研究,提出了“返还型”老年防癌疾病保险的期权定价框架:将保险合同看作是每一份期权产品中附带了多个具有约定敲出条件的看涨障碍期权,通过蒙特卡洛方法模拟出癌症发生率随时间推移可能呈现的变化路径,分路径计算其蕴含的结构化期权产品的价格。(3)结合定性和定量研究的方法,从数据风险、模型风险和外部环境风险等三方面分析老年防癌疾病保险的定价风险,并提出针对性的风险管理措施。更多还原

【Abstract】 The latest national statistics show that by the end of 2021,14.2% of China’s elderly aged 65 and above,China has entered a "deeply aging society" ahead of schedule,and health protection for the elderly has become a big issue facing the current society.In the context of our government’s large burden of medical care and health management for the elderly and the increasing burden of medical expenses,commercial health insurance,which is a strong protection for the health risks of the elderly,is playing an increasingly important role.The insurance industry has been thinking about and exploring how to develop health insurance that is more suitable for the elderly and solve the health protection problems of the elderly population.2013 saw the launch of the first elderly cancer disease insurance in China,filling the gap in the elderly health insurance market in China and improving the level of health protection for all.The probability of disease occurrence and its loss distribution are the quantitative basis of health insurance actuarial calculations,but there are few studies on disease occurrence measurement and health insurance rate determination in China,and the lack of relevant research results has hindered the further development of health insurance actuarial techniques.Especially for the elderly cancer insurance,the target population is the elderly with high incidence of diseases,the incidence of cancer is high,volatile,and the future development trend is highly uncertain,so the importance of reasonable basic data and proper actuarial pricing techniques is especially high,which is directly related to the sustainable and stable development of these products in China’s health insurance market.Therefore,the study of cancer incidence measurement and senior citizen cancer insurance pricing has important theoretical significance and practical value.On the basis of summarizing the theories of disease incidence research and disease insurance pricing at home and abroad,this paper researches the measurement of cancer incidence in the elderly population,establishes a dynamic stochastic model suitable for cancer incidence in the elderly population in China,and proposes a new pricing method for elderly cancer disease insurance,reflecting the uncertainty of the trend of cancer incidence due to time changes in this pricing method.The details of this paper are as follows.(1)Based on the dynamic and static characteristics of cancer incidence data of people aged 50-75 in China,six stochastic dynamic models such as the Lee-Carter model are introduced to measure cancer incidence and its future changes,and the one with the best fit is selected to measure the cancer disease risk of the elderly population.(2)Based on the summary and research of three traditional actuarial pricing methods,such as the Mann-Lien method,we propose an option pricing framework for "refundable" old-age cancer insurance: the insurance contract is considered as each option product with multiple call-barrier options with agreed knock-out conditions,and the path of possible changes in cancer incidence over time is simulated by the Monte Carlo method.The paths of possible changes in cancer incidence over time are modeled by Monte Carlo methods,and the prices of the structured option products embedded in them are calculated in sub-paths.(3)Combining qualitative and quantitative research methods,we analyze the pricing risk of senior cancer disease insurance from three aspects,including data risk,model risk,and external environment risk,and propose targeted risk management measures.更多还原