考虑交易成本和基数约束的可能性均值-下半方差多期投资组合模型  

A possible mean-semi-variance muti-period portfolio model by considering transaction costs and cardinality constraints

作　　者：胡晨阳 高岳林 孙滢[2] HU Chen-yang;GAO Yue-lin;SUN Ying(School of Mathematics and Information Science, North Minzu University, Yinchuan 750021, Ningxia, China;Ningxia Collaborative Innovation Center of Scientific Computing and Intelligent Information Processing, North Minzu University, Yinchuan 750021, Ningxia, China;Ningxia Key Laboratory of Intelligent Information and Big Data Processing, North Minzu University, Yinchuan 750021, Ningxia, China)

机构地区：[1]北方民族大学数学与信息科学学院,宁夏银川750021 [2]北方民族大学宁夏科学计算与智能信息处理协同创新中心,宁夏银川750021 [3]北方民族大学宁夏智能信息与大数据处理重点实验室,宁夏银川750021

出　　处：《宝鸡文理学院学报（自然科学版）》2022年第1期21-26,共6页Journal of Baoji University of Arts and Sciences(Natural Science Edition)

基　　金：国家自然科学基金项目(11961001);宁夏高等教育一流学科教育基金(NXYLXK2017B09);北方民族大学重大专项资助(ZDZX201901)。

摘　　要：目的提出一个改进的差分进化算法,求解在模糊环境下建立的可能性均值-下半方差多期投资组合模型。方法考虑模糊环境下多个摩擦因素,利用交易限制引入资产的基数约束,构建一个可能性均值-下半方差多期投资组合模型。在标准的DE算法中引入levy飞行随机游走策略增强算法寻优能力。结果与结论数值实验和仿真结果表明,利用改进的DE算法得到的投资收益比标准的DE算法高,且风险值比标准的DE算法低,同时也验证了建立的模型的合理性和实用性。Purposes—To solve the probability mean-lower-semi-variance multi-period portfolio model established in the fuzzy environment with an improved differential evolution algorithm.Methods—Considering multiple friction factors in the fuzzy environment,a possibility mean-lower-semi-variance multi-period portfolio model is constructed by introducing the asset base constraints with transaction restriction.The levy flight random walk strategy is introduced into the standard DE algorithm to enhance the optimization ability of the algorithm.Result and Conclusion—Numerical experiments and simulation results show that the improved DE algorithm has a higher return on investment than the standard DE algorithm,and a lower risk value than the standard DE algorithm,which also verifies the rationality and practicability of the established model.

关 键 词：投资组合 基数约束 模糊环境 levy飞行策略 差分进化算法 

分 类 号：F380[经济管理—产业经济]