我国基本养老保险制度中的长寿风险精算管理研究

随着人口平均寿命的逐步延长，长寿风险已成为一种日益严重的社会风险，长寿风险对我国基本养老保险制度的安全稳定运行提出了严峻的挑战。针对我国基本养老保险制度中的长寿风险问题，从长寿风险识别角度，结合中国人口死亡率数据的特点，构建基于数据差异的Lee-Carter-最小叉熵死亡率模型。从风险评估角度，研究度量长寿风险的精算指标模型及基于GlueVaR的评估方法。从风险控制角度，针对社会统筹，综合人口红利、养老保险基金平衡及个人退休满意度等三个层面，研究长寿风险暴露下的延迟退休策略、构建度量延迟退休效果的集结算子评价模型；针对个人帐户，研究长寿风险暴露下的个人账户养老金动态投资策略、应对长寿风险的个人账户月养老金动态计发数模型。从风险转移角度，研究构建生存指数、分层长寿债券的设计及定价模型。通过上述研究，以期为应对基本养老保险制度中的长寿风险，促进国民经济健康发展做出贡献。

With the increasing of life expectancy, longevity risk has become a seriously social risk, and it is severely throating the fundamental pension system of our China. Based on this condition, we try to solve this problem via the following approaches. From the view point of longevity risk identification, combing with the feature of population mortality data in China, we will construct a mortality forecasting model, i.e., Lee-Carter-minimization cross entropy with the difference of data. From the point of risk assessment, we provide some actuarial indexes and an approach based on GlueVaR to measuring the longevity risk. From the point of risk controlling, according to the social pooling account, we put forward to a retirement strategy by considering demographic dividend, fundamental pension system of our China and individual retirement decision-making satisfaction degree, and develop an aggregation operator model for measuring the effect of delaying retirement alternatives. According to the individual account, we provide the dynamic portfolio selection based on rewards and punishment coefficient under the exposure of longevity risk, in addition, we put forward to a dynamic coefficient model for measuring the monthly pension in individual account. From the angle of risk transfer, according to the fundamental pension plan, we establish a survival index model and investigate the design and pricing model of hierarchy longevity bond. It is hoped that the above mentioned work can decrease the pressure of fundamental pension system to the longevity risk and can make contribution to the health development of China economic.