摘要
分析了最优化投资组合不稳定的原因,在此基础上,推广了由前人提出的一个新方法,即根据随机矩阵定理(RMT)对样本协差阵进行变化以得到一个更好的投资方案.首先对该方法中未证明的部分给出了更充分的实证理由,并运用这种思想,结合中国证券市场的实际数据,来模拟"真实的投资".而这种方法得到的投资方案,较之传统方法有较好的收益率和方差,且可被前人对风险预测的分析方法所解释.
The reasons why the risk of optimal portfolio is unstable was analyzed.Based on this analysis a new method was tried,which was proposed by the previous researchers,to dispose the sample covariance matrix based on random matrix theory（RMT）,to get a better result.Generally,we first give more sufficient and practical evidence to support the method’s reasonableness and power,then apply this idea to simulate the＂real investment＂,using real data of the Chinese stock market.It is found that the results have better mean return and stability than those of the traditional method,and can be explained by the analysis of the prediction to the risk proposed by the previous researchers.
