计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (8): 238-243.DOI: 10.3778/j.issn.1002-8331.1807-0294

• 工程与应用 • 上一篇    下一篇

基于神经网络集成学习股票预测模型的研究

谢  琪1,程耕国1,徐  旭2   

  1. 1.武汉科技大学 信息科学与工程学院,武汉 430065
    2.格勒诺布尔高等商学院 高等商业学院,格勒诺布尔 38000
  • 出版日期:2019-04-15 发布日期:2019-04-15

Research Based on Stock Predicting Model of Neural Networks Ensemble Learning

XIE Qi1, CHENG Gengguo1, XU Xu2   

  1. 1.School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan 430065, China
    2.Ecole Supérieure de Commerce, Grenoble Ecole de Management, Grenoble 38000, France
  • Online:2019-04-15 Published:2019-04-15

摘要: 基于深度学习的原理构建出六层长短记忆神经网络,通过集成学习中Bagging方法组合8个长短记忆神经网络。使用基于神经网络集成学习模型预测中国人民币普通股市场。实验测试了从2012年1月4日到2017年12月29日这期间的上海证券综合指数、深圳证券综合指数、上证50指数、沪深300指数、中小企业板指数和创业企业板指数。实验结果为模型的准确率达到58.5%,精确率为58.33%,召回率为73.5%,[F1]值为64.5%,AUC值为57.67%,取得了较好的预测效果。

关键词: 长短记忆神经网络, 装袋算法, 股票, 准确率, 精确率

Abstract: Based on deep learning, six layer long short-term memory neural networks are constructed. Eight long short-term memory neural networks are combined with bagging method in ensemble learning and predicting model of neural networks ensemble learning is used in Chinese Stock Market. The experiment tests Shanghai Composite Index, Shenzhen Composite Index, Shanghai Stock Exchange 50 Index, Shanghai-Shenzhen 300 Index, Medium and Small Plate Index and Gem Index during the period from January 4, 2012 to December 29, 2017. The model’s accuracy is 58.5%, precision is 58.33%, recall is 73.5%, [F1] value is 64.5%, and AUC value is 57.67%, which reflect a good prediction outcome.

Key words: Long Short-Term Memory(LSTM), bagging, stock, accuracy, precision