Computer Engineering and Applications ›› 2016, Vol. 52 ›› Issue (18): 167-172.

Previous Articles     Next Articles

Stock market selection model based on genetic algorithm

RONG Rong, WU Ping   

  1. School of Computer Science and Software Engineering, East China Normal University, Shanghai 200062, China
  • Online:2016-09-15 Published:2016-09-14

基于遗传算法的股票市场选择模型

戎  容,吴  萍   

  1. 华东师范大学 计算机科学与软件工程学院,上海 200062

Abstract: In order to increase the income of investors in the stock market and solve the important issue about stock selection in the securities investment, a stock selection model based on genetic algorithm is proposed. This method uses financial indicators of listed companies as characteristics of samples. Firstly, in order to overcome the instability of K-means algorithm, it doses the cluster analysis on stocks in the same field using the K-means algorithm based on genetic algorithm. Then stocks in the poor class should be removed. Secondly, it encodes screening conditions. An improved genetic operator is proposed to solve the premature convergence of traditional genetic algorithm in dealing with complex problems. Then it uses the improved genetic algorithm to find parameters of the stock selection model which can maximize the income of the investment in the stock market. Experimental results demonstrate that this method has a great effect on stock selection and is valuable to investors.

Key words: stock selection, genetic algorithm, cluster analysis, investment decision, combinatorial optimization

摘要: 为提高投资者在股票市场的收益,解决在证券投资中股票选择这一重要问题,提出一种基于遗传算法的股票选择模型。算法以上市公司的财务指标为样本特征,为克服K-means算法的不稳定性,采用基于遗传算法的K-means算法对同一板块股票进行聚类分析,剔除财务指标较差的一类中的股票。对筛选条件编码,为解决传统遗传算法处理复杂问题时存在的过早收敛现象,提出改进的遗传算子,利用改进的遗传算法寻找使股票市场投资收益最大化的选股模型参数。实验结果表明,该算法在股票选择上具有较好的效果,可供市场投资者借鉴。

关键词: 股票选择, 遗传算法, 聚类分析, 投资决策, 组合优化