改进的人工蜂群算法在投资组合中的应用研究  

Application Research on Improved Artificial Bee ColonyAlgorithms in Portfolio Selection

作　　者：邓雪[1] 方雯 DENG Xue;FANG Wen(School of Mathematics,South China University of Technology,Guangzhou 510640,China)

机构地区：[1]华南理工大学数学学院,广东广州510640

出　　处：《数学的实践与认识》2023年第8期1-12,共12页Mathematics in Practice and Theory

基　　金：国家社会科学基金(21BTJ069)。

摘　　要：基于Markowitz投资组合模型,构建了带有持有量和交易量限制的新的均值-方差模型.在标准人工蜂群算法中Deb选择策略的基础上,受粒子群算法思想的启发,对其初始化阶段及搜索算子进行修正,提出了改进的人工蜂群算法.同时,使用改进人工蜂群算法求解所构建的模型,并将计算结果与标准人工蜂群算法及粒子群算法的计算结果进行比较.结果表明:改进的人工蜂群算法结合了标准人工蜂群算法以及粒子群算法的优点,求得的结果具有更好的稳定性及收敛性,更加精准.最后,比较了两个投资模型的有效前沿,通过分析得出:限制条件会使投资组合的整体收益率下降;与风险厌恶者相比,限制条件对于风险偏好型投资者的影响更大.Based on Markowitz portfolio model,a new mean-variance model with restrictions on holdings and trading volumes is proposed in this paper.On the basis of Deb selection strategy in standard artificial bee colony algorithm,inspired by the idea of particle swarm optimization,the initialization stage and search operator are modified,and then an improved artificial bee colony algorithm is proposed.At the same time,this improved artificial bee colony algorithm is used to solve the constructed two models,and the results are compared with those of standard artificial bee colony algorithm and particle swarm optimization algorithm.The results show that the improved artificial bee colony algorithm combines the advantages of standard artificial bee colony algorithm and particle swarm optimization algorithm,and the results obtained have better stability,convergence and accuracy.Finally,the effective frontiers of the two portfolio models are compared.Through analysis,it is concluded that the restriction conditions will reduce the overall return of the portfolio;compared with risk-averse investors,the restriction conditions have a greater impact on risk-seeking investors.

关 键 词：投资组合 Deb选择策略 改进的人工蜂群算法 粒子群算法 有效前沿面 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] F830[自动化与计算机技术—控制科学与工程]