基于捕食策略的粒子群算法求解投资组合问题

发布时间：2018-02-15 15:58

本文关键词： 捕食搜索策略粒子群算法投资组合模型　出处：《暨南大学》2013年硕士论文　论文类型：学位论文

【摘要】：本文的主要内容是总结群智能优化算法的优缺点的基础上,提出了一种基于捕食策略的粒子群优化算法，然后将该算法与标准粒子群优化算法(ParticleSwarm Optimization, PSO)，通过测试函数进行比较，检验该算法能够有效的提高收敛速度和精度，最后将该算法应用到求解证券投资组合优化问题中。现代投资组合理论就是解决最大投资收益与避免风险之间的基本权衡关系的学说，每个理性的投资者都在追求在给定的收益下希望风险最小，或者在风险一定的条件下收益最大。本文从应用角度出发，通过分析中国证券市场现实投资环境和实际特点，建立了一个考虑完整费用的证券投资组合模型。并针对标准粒子群算法容易陷入局部最优和搜索精度不高的缺点，提出了基于捕食策略的粒子群优化算法（PSPSO），将其用于求解投资组合模型。捕食搜索策略可以通过调节限制级别来控制粒子群的搜索空间，从而平衡全局搜索和局部搜索。最后采集实际证券投资数据进行数值模拟，利用基于捕食策略的粒子群优化算法对模型进行求解，数值模拟结果显示，与基本粒子群优化算法相比，，本文提出的基于捕食策略的粒子群优化算法能够有效的提高收敛速度以及收敛精度。
[Abstract]:The main content of this paper is the basis of summing up the advantages and disadvantages of swarm intelligence optimization algorithms and proposes a foraging strategy based on particle swarm optimization algorithm, then the algorithm and standard particle swarm optimization algorithm (ParticleSwarm, Optimization, PSO), were compared by test function, test the algorithm can improve the convergence speed and the accuracy of the algorithm is used to solve the portfolio optimization problem.
The modern portfolio theory is the theory of solving the basic trade-off between maximum investment income and avoiding risk. Every rational investor is pursuing the minimum risk under given income or the maximum profit under certain risk.
This paper from the application point of view, through the analysis of the securities market Chinese actual investment environment and characteristics, we considered a complete cost model of portfolio investment. According to the standard particle swarm algorithm is easy to fall into local optimum and the search accuracy, proposed foraging strategy based on particle swarm optimization algorithm (PSPSO), the to solve the portfolio investment model. Predatory search strategy can control the particle swarm search space by adjusting the limit level, so as to balance the global search and local search.
Finally, the actual acquisition of securities investment data by using numerical simulation to solve the model predation strategy based on particle swarm optimization algorithm, simulation results show that, compared with the basic particle swarm optimization algorithm, this paper proposed the foraging strategy of particle swarm optimization algorithm can improve the convergence speed and convergence accuracy based on effective.

【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP18;F830.91

【参考文献】