Evolved Neural Network Based Intelligent Trading System for Stock Market

摘要：

　　In the present study, evolved neural network is applied to construct a new intelligent stock trading system.First, heterogeneous double populations based hybrid genetic algorithm is adopted to optimize the connection weights of feedforward neural networks.Second, a new intelligent stock trading system is proposed to generates buy and sell signals automatically through predicting a new technical indicator called medium term trend.Compared to traditional NN,the new model provides an enhanced generalization capability that both the average return and variance of performance are significantly improved.

关键词： genetic algorithm neural network network training stock trading system

作者: Lifeng Zhang Yifan Sun

作者单位: School of Information, Renmin University of China 59, Zhongguancun Street, Haidian, Beijing, P.R.China, 100872

会议类型: 国际会议

会议名称: 4th international Conference,ICSI2013(第4届群体智能国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 478-488

在线出版日期: 2013-06-12（万方平台首次上网日期，不代表论文的发表时间）