数据驱动的扭曲均值-半方差投资组合选择  

Mean-Semivariance Portfolio Selection with Distortion Based on a Data Driven Approach

作　　者：杨东方 李孟雨 王天放 米辉[1] 刘国祥[1] Yang Dongfang;Li Mengyu;Wang Tianfang;Mi Hui;Liu Guoxiang(School of Mathematical Sciences,Nanjing Normal University,Nanjing 210023,China)

机构地区：[1]南京师范大学数学科学学院,江苏南京210023

出　　处：《南京师大学报（自然科学版）》2023年第2期7-14,共8页Journal of Nanjing Normal University(Natural Science Edition)

基　　金：国家自然科学基金项目(61304065).

摘　　要：运用样本平均近似的数据驱动方法研究了带概率扭曲的均值-半方差投资组合优化模型,结合经典的遗传算法(ICA)和帝国竞争算法(GA),提出了ICA-GA混合算法.利用真实市场数据,对模型进行实证分析并求解有效前沿.最后,通过比较算法程序运行时间,表明本文创新的ICA-GA混合算法融合了帝国竞争算法和遗传算法两者的优势,比它们有更好的表现.This paper studies a mean-semivariance portfolio optimization problem with probability distortion by using the data driven sample average approximation(SAA)approach.The paper builds ICA-GA hybrid algorithm based on the traditional imperial competitive algorithm(ICA)and genetic algorithm(GA).Employing the true market data,the model is empirically analyzed and the effective frontier is solved.Finally,by comparing the running time of the computer programs,it shows that the ICA-GA hybrid algorithm in this paper combines the advantages of imperial competitive algorithm and genetic algorithm,and it performs better than them.

关 键 词：带概率扭曲的均值-半方差 投资组合优化 ICA-GA混合算法 数据驱动方法 

分 类 号：O211[理学—概率论与数理统计] F830[理学—数学]