摘要
本文用可容许误差来处理金融市场中的不确定性。首先,将资产的预期收益和风险都用可容许误差来表示,在考虑交易成本、借贷约束、基数约束和上下界等约束条件的基础上,提出具有基数约束的可容许均值-方差投资组合模型。然后,引入情感系数λ来描述投资者对未来资本市场的预期收益。由于该模型是一个非线性整数规划问题,只能采用智能算法进行求解。因此,本文设计了一种遗传算法求解,并通过Cplex进行可行性分析。最后,对不同的情感系数、借贷约束、上下界约束的有效前沿进行分析,从而验证了模型和遗传算法的有效性。
In this paper,we use admissible errors to deal with uncertainty in financial markets.Firstly,the expected return and risk of assets are expressed by admissible errors.Considering the transaction costs,borrowing constraints,cardinality constraints and threshold constraints,we present a new admissible mean-variance portfolio selection model with cardinality constraints.Then,the emotional coefficientλis introduced to describe the expected return of the investors on the capital market in the future.Since the proposed model is a nonlinear integer programming problem,it can only be solved by intelligent algorithms.Therefore,we design a genetic algorithm,and conduct the feasibility analysis by Cplex.Finally,the effective frontiers comparison analyses of different emotional coefficients,borrowing constraints and threshold constraints are given to examine the model and the designed genetic algorithm.
作者
党世力
张鹏
李璟欣
DANG Shi-li;ZHANG Peng;LI Jing-xin(School of Economics&Management,South China Normal University,Guangzhou 510006,China)
出处
《模糊系统与数学》
北大核心
2023年第4期92-103,共12页
Fuzzy Systems and Mathematics
基金
国家自科项目(71271161)
广东省社科项目(GD19CGL32)
广东省软科学项目(2019A101002052)
广东省软科学项目(2019A101002066)
关键词
可容许投资组合
均值-方差
基数约束
遗传算法
情感系数
Admissible Portfolio Selection
Mean-variance
Cardinality Constraint
Genetic Algorithm
Emotional Coefficient