通过随机矩阵方法识别、消除极端或非无关抽样样本,提高Markowitz模型的参数估计精度,改进应用Markowitz模型的效果;同时,针对抽样不足的情况,使用Bootstrap方法较好地解决了该问题.
Markowitz's mean-variances model in this paper is improved,and the random matrix theory is used that can identify extreme sampling data and relevance data to get rid of those data such that more accurate estimate of mean and variance can be gotten.Then Bootstrap method to solve the problem of insufficient sample is used.