›› 2011, Vol. 24 ›› Issue (8): 15-.

• 论文 • 上一篇    下一篇

改进的BP算法在股市预测中的应用

冯居易   

  1. (西安财经学院 信息学院,陕西 西安 710100)
  • 出版日期:2011-08-15 发布日期:2011-08-19
  • 作者简介:冯居易(1979—),女,硕士,讲师。研究方向:数据挖掘,金融预测。
  • 基金资助:

    陕西省教育厅科学研究计划基金资助项目(2010JK554);陕西省自然科学基金资助项目(2007F25)

Application of Improved BP Algorithm in Stock Market Prediction

 FENG Ju-Yi   

  1. (School of Information,Xi'an University of Finance and Economics,Xi'an 710100,China)
  • Online:2011-08-15 Published:2011-08-19

摘要:

股票价格预测是证券界和学术界的一个重要的研究课题。神经网络具有强大的非线性逼近能力,文中采用MO-VLBP神经网络建立股票价格预测模型,对某银行的收盘价进行预测。实验结果证明,MO-VLBP网络模型应用于股票价格的短期预测,运算速度快、预测精度高。

关键词: 神经网络, 动量-可变学习率BP算法, 股票价格, 预测模型

Abstract:

Stock price prediction is an important research topic of the securities industry and academia.Neural network has a strong power of non-linear approximation.The author utilizes MO-VLBP neural network to build the prediction model for Beijing Bank's closing price prediction.The experiment proves that the MO-VLBP prediction model applied into the stock price forecasting has the advantages of fast computing speed and good forecasting precision,and has a value of promotion and application.

Key words: neural networks;MO-VLBP algorithm;stock price;prediction

中图分类号: 

  • TP391