CVaR约束下的鲁棒投资组合模型

Robust portfolio selection using CVaR strategy

针对均值-CVaR投资组合模型的解对其中参数变化敏感的问题,构造形式较为简单的"势"不确定集作为对模型中不确定参数取值的近似,由此给出的鲁棒模型易于求解,且得到的解兼顾鲁棒性和最优性,同时保持了原问题的计算难度.应用实际交易数据对所提出的模型进行数值实验和比较,结果表明,此模型能够获得较好的财富增长率的投资策略,并能有效地分散最优投资组合的风险.

In view of the solution of the average-CVaRportfolio model is sensitive to the change of the parameters,the＂cardinality＂uncertainty set which is relatively simple was structured as the approximation to the uncertain parameters.Then the corresponding robust form was easy to solve,and the solution was both robust and optimal,while the original calculating difficulty of the problem was maintained.The empirical analysis and comparisons from the real market data indicate that the proposed model can obtain a portfolio strategy with the better wealth growth rate and diversify the risk of the optimal portfolio efficiently.