基于TanW-ArccosC PSO算法的PP求解与实证分析
Empirical Analysis and Solution of Portfolio Problem Based on TanW-ArccosC PSO Algorithm
姚琳 1贾文生1
作者信息
- 1. 贵州大学数学与统计学院,贵州 贵阳 550025;贵州省博弈决策与控制重点实验室,贵州 贵阳 550025
- 折叠
摘要
通过结合正切函数Tan-W和反余弦函数Arccos-C提出了一种改进的粒子群优化算法,简称TanW-ArccosC PSO算法.TanW-ArccosC PSO算法通过对惯性权重和学习因子的改进,增加了粒子群的多样性,增强了算法的搜索能力,提高了算法的收敛速度.针对投资组合问题,通过在大智慧软件中随机提取数据,利用 MATLAB 软件,分别用改进的 TanW-ArccosC PSO算法和标准PSO算法进行求解与实证分析其投资组合问题的投资比例和CVaR值,实证分析结果表明TanW-ArccosC PSO算法具有更良好的搜索能力、低风险性以及可操作性.
Abstract
In this paper,an improved particle swarm optimization algorithm,abbreviated as TanW-ArccosC PSO algorithm,is proposed by combining the tangent function Tan-W and the inverse cosine function Arccos-C.The TanW-ArccosC PSO algorithm increases the diversity of particle swarms,enhances the search capability of the algo-rithm,and improves the convergence speed of the algorithm by improving the inertia weights and learning factors.For the portfolio problem,this paper used the improved TanW-ArccosC PSO algorithm and the standard PSO algorithm to solve and empirically analyze the investment ratio and CVaR value of the portfolio problem by randomly extracting the data in the Dazhihui software and using matlab software.The empirical analysis results show that the TanW-ArccosC PSO algorithm has better search ability,low risk and operability.
关键词
粒子群算法/惯性权重/学习因子/投资组合问题Key words
Particle swarm optimization/Inertia weights/Learning factors/Portfolio Problems(PP)引用本文复制引用
基金项目
国家自然科学基金(12061020)
贵州省科技基金会项目(20201Y284)
贵州省科技基金会项目(20205016)
贵州省科技基金会项目(2021088)
出版年
2024
NETL
NSTL
万方数据