中文摘要：

研究了马尔可夫机制转换模型下确定缴费型养老金计划的最优投资问题.假定市场中风险资产价格与企业员工的工资都满足马尔可夫调制的几何布朗运动模型,它们的预期回报率和波动率都依赖于市场经济状态,其经济状态由一连续时间马尔可夫链来描述.利用最终财富的最大期望效用准则,得到了养老金管理者的最优投资策略,结果表明市场的经济状态对最优投资策略有着很大的影响.最后通过数值计算分析了市场利率和绝对风险厌恶系数与最优投资策略的关系.

英文摘要：

The problem of optimal investment for a defined contribution（DC） pension plan under a Markov-regime switching model is studied.The price dynamics of a risky asset and the total salary of the company staffs are assumed to follow a Markov-modulated geometry Brownian motion,their appreciation rates and volatilities depend on the states of an economy,which are described by a continuous time Markov chain.In the end,an optimal investment strategy of a pension manager is obtained by maximize the expected utility of terminal wealth.The results indicate that the states of an economy have a significant effect on the optimal investment strategies,and a numerical analysis is provided which illustrates the effect of the market interest rate and absolute risk aversion parameter on the optimal investment strategy.