东北大学学报(自然科学版) ›› 2013, Vol. 34 ›› Issue (6): 901-904.DOI: -

• 管理科学 • 上一篇    下一篇

基于IDNPSO-BP神经网络的股票市场指数预测

刘家和1,金秀1,陈露艳2,苑莹1   

  1. (1.东北大学工商管理学院,辽宁沈阳110819;2.中国人民大学财政金融学院,北京100872)
  • 收稿日期:2012-10-25 修回日期:2012-10-25 出版日期:2013-06-15 发布日期:2013-12-31
  • 通讯作者: 刘家和
  • 作者简介:刘家和(1987-),男,广东佛山人,东北大学博士研究生;金秀(1963-),女,辽宁沈阳人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(70901017,71271047);中央高校基本科研业务费专项资金资助项目(N100406003).

Stock Market Index Forecasting Based on IDNPSOBP Neural Network

LIU Jiahe1, JIN Xiu1, CHEN Luyan2, YUAN Ying1   

  1. 1. School of Business Administration, Northeastern University, Shenyang 110819, China; 2. The School of Finance, Renmin University of China, Beijing 100872, China.
  • Received:2012-10-25 Revised:2012-10-25 Online:2013-06-15 Published:2013-12-31
  • Contact: LIU Jiahe
  • About author:-
  • Supported by:
    -

摘要: 针对动态邻居粒子群算法的局限性,引入新的动态邻居拓扑结构,动态调整粒子群算法参数设置,提出改进的动态邻居粒子群算法(IDNPSO).为了提高BP神经网络模型的预测准确性,提出一种基于改进动态邻居粒子群算法的BP神经网络模型(IDNPSO-BP神经网络).利用IDNPSO-BP神经网络和GA-BP神经网络对上证指数、深证指数进行预测,结果表明IDNPSO-BP神经网络的预测误差优于GA-BP神经网络,具有股票市场指数预测能力.

关键词: 神经网络, 动态邻居, 粒子群算法, 市场指数, 预测

Abstract: An improved dynamic neighborhood particle swarm optimization (IDNPSO) was proposed. A new topological structure which constructs dynamic neighbors was induced, and the parameter settings were dynamically adjusted. To improve the predictive accuracy of BP neural network, an improved prediction method of optimized BP neural network based on IDNPSO was introduced. Making use of the index price of Shanghai composite index and Shenzhen composite index, comparison of the forecast performance between IDNPSOBP and GABP neural networks was taken. The result showed that INDPSOBP neural network outperformed GABP neural network, and had the ability to forecast stock index price.

Key words: neural network, dynamic neighborhood, particle swarm optimization, stock index, forecasting

中图分类号: