约束粒子群算法求解自融资投资组合模型研究  

Constrain Particle Swarm Optimizer for Solving Sel-Financing Portfolio Model

作　　者：刘衍民[1,2] 赵庆祯[2] 牛奔[3] 

机构地区：[1]遵义师范学院数学系,贵州遵义563002 [2]山东师范大学管理与经济学院,山东济南250014 [3]深圳大学管理学院,广东深圳518060

出　　处：《数学的实践与认识》2011年第2期78-84,共7页Mathematics in Practice and Theory

基　　金：国家自然科学基金(71001072);贵州教育厅社科项目(0705204);遵义师院课题(2007018;基07015;07017)

摘　　要：在马克维茨投资组合的均值-方差模型框架下,给出限制投资数量的自融资投资组合优化模型.在金融市场上有广泛应用,为了有效地求解此类问题的最优解,采用一种基于广义学习策略的约束粒子群算法(CPSO).CPSO算法具有广义的学习策略,极大地提升了种群的多样性,进而提升种群跳出局部最优解的能力.在基准函数测试中,结果显示CPSO算法有较好的运行结果.在自融资投资组合优化模型上,优化结果表明CPSO算法是可行的,有效的,并有较好的优化结果.Based on Markowitz＇s mean-variance portfolio model,a self-financing portfolio optimal model with constraints of invest proportions is proposed,which is widely used in financial market.In order to effectively solve the optimal solution of this model,a comprehensive learning constraint PSO algorithm is proposed（CPSO for short）.In CPSO,the comprehensive learning strategy is adopted,which greatly improve the diversity of the swarm and improve the ability to escape from local optima.The experiments on benchmarks indicate that the proposed algorithm has good performance.And in self-financing portfolio optimal model,the CPSO algorithm is feasible and effective,and the more effective results.

关 键 词：约束 粒子群算法 投资组合 自融资 

分 类 号：F224[经济管理—国民经济] F830.9