论文部分内容阅读
传统Mean—CVa R模型在实际运用中,其最优解在投资组合优化过程中容易产生过多的微小权重,导致最优投资组合中非零权重的个数非常多。同时,还存在大权重过大问题,使风险不能有效地分散化。实证研究证明,应用范数正则化理念,并结合传统Mean—CVa R,提出并采用基于稀疏优化的Mean—CVa R模型,对解决大数据市场背景下的投资组合优化问题具有一定的指导意义。
In the practical application, the traditional Mean-CVa R model tends to generate too many small weights in the optimal portfolio optimization process, resulting in a very large number of non-zero weights in the optimal portfolio. At the same time, there is still a problem of over-weighting so that risks can not be effectively decentralized. The empirical research proves that applying the norm regularization concept and combining with the traditional Mean-CVa R, it is instructive to propose and adopt the Mean-CVa R model based on sparse optimization to solve the portfolio optimization problem in the big data market.