Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (11): 3336-3341.DOI: 10.11772/j.issn.1001-9081.2018040742

Previous Articles     Next Articles

Financial time series prediction by long short term memory neural network with tree structure

YAO Xiaoqiang1, HOU Zhisen2   

  1. 1. College of Air and Missile Defense, Air Force Engineering University, Xi'an Shaanxi 710051, China;
    2. School of Software Engineering, Xi'an Jiaotong University, Xi'an Shaanxi 710049, China
  • Received:2018-04-12 Revised:2018-06-14 Online:2018-11-10 Published:2018-11-10

基于树结构长短期记忆神经网络的金融时间序列预测

姚小强1, 侯志森2   

  1. 1. 空军工程大学 防空反导学院, 西安 710051;
    2. 西安交通大学 软件学院, 西安 710049
  • 通讯作者: 姚小强
  • 作者简介:姚小强(1985-),男,河南宜阳人,讲师,硕士,主要研究方向:智能信息处理;侯志森(1982-),男,辽宁海城人,硕士研究生,主要研究方向:机器学习、数据挖掘、量化投资。

Abstract: Aiming at the problem that traditional methods can not effectively predict multi-noise and nonlinear time series, focusing on multi-scale features fusion, a prediction method based on tree structure Long Short-Term Memory (LSTM) neural network was proposed and verified. First of all, the core methods of realizing the prediction were proposed, and the internal advantages of the methods were analyzed. Secondly, the prediction model based on tree structure LSTM neural network was constructed. Finally, the model was verified based on the international gold spot transaction data of the last ten years. The results show that the prediction accuracy is nearly 10 percentage points higher than the minimum success rate, and the availability of the methods is proved.

Key words: tree structure, Long Short-Term Memory (LSTM) neural network, financial time series, prediction

摘要: 针对传统方法对多噪声、非线性的时间序列无法进行有效预测的问题,以多尺度特征融合为切入点,提出并验证了基于树结构长短期记忆(LSTM)神经网络的预测方法。首先,提出了实现预测目标的核心方法,并分析了方法的内在优势;其次,构建了基于树结构长短期记忆神经网络的预测模型;最后,基于最近十年的国际黄金现货交易数据对模型进行了验证。实验结果表明,所提算法预测准确率高出最小成功率近10个百分点,证实了所提方法的有效性。

关键词: 树结构, 长短期记忆神经网络, 金融时间序列, 预测

CLC Number: