基于VaR控制下的动态优化投资组合

Dynamic optimal portfolio in a VaR framework

本文应用的动态优化投资组合模型是在VaR的约束下,调整投资组合的配置,使期望收益达到最大。基于投资组合中每一种资产的收益率序列,模型在不断变化的数据窗口下,首先估计模型参数,然后求解优化模型,得到每日投资组合中风险资产在VAR约束下的最优配置和借贷比率。这种方法对构筑新的风险资产投资组合的决策,以及对已有投资组合中资产配置的优化具有重要的指导意义。选取中国A股市场的4只股票,在收益率服从正态分布的假定下,确定投资组合中的资产配置以及借贷比率,并且讨论了模型参数的敏感性。

A dynamic portfolio model that maximizes expected returns subject to a Value-at-Risk constraint by adjusting the set of portfolio weights, has been investigated. Parameters of the model were first investigated by several methods based on the return serials of each financial asset in the portfolio along with the movement of the data windows. Then, by solving for the optimal model, the best daily allocations of the optimal portfolio and the rate of borrowing and lending subject to a Value-at-Risk constraint were Obtained. The results are of value in setting up a new risk portfolio and optimizing the daily portfolio allocations. Four shares in the Chinese A stock market were selected and, with the assumption of a normal distribution of the returns, the daily portfolio allocations and the rate of borrowing and lending were obtained. The sensitivity of the model parameters are also discussed in this paper.