关系驱动的基于时间卷积网络的股票走势预测算法
Relation-Driven Stock Trend Prediction Model Based on Time Convolutional Network

DOI: 10.12677/CSA.2021.1110259, PP. 2555-2567

Keywords: 图卷积网络，时间卷积网络，小波包变换，股票价格预测
Graph Convolution Network, Time Convolution Network, Wavelet Packet Transform, Stock Price Prediction

Full-Text   Cite this paper   Add to My Lib

Abstract:

由于股票市场的复杂性和波动性，股票价格走势预测被认为是一项具有挑战性的任务。基于图卷积网络(Graph Convolution Network, GCN)处理非欧式结构数据的能力，本文提出了一种股票走势预测算法。该算法通过时间卷积网络(Time Convolution Network, TCN)从股票的历史交易数据中提取特征，使用图结构对股票之间的关系进行量化，应用图卷积网络从相关股票中学习局部特征表示来预测股价走势。与以往的研究不同，本文强调个体特征和多尺度局部特征的融合，并应用小波包变换对数据进行去噪以提高算法的性能。本文提出的算法在中国大陆沪深300成分股数据集中进行了验证，在两个指标上取得了优于常见时间序列预测基线模型2%和3%的性能表现。
Stock price movement prediction is considered a challenging task due to the complexity and volatility of the stock market. Based on the ability of graph convolution network (GCN) to process non-European structured data, this paper proposes a stock trend prediction algorithm. The algorithm extracts features from the historical trading data of stocks through TCN (Time Convolution Network), quantifies the relationship between stocks using graph structure, and uses GCN to learn local feature representation from related stocks to predict the movement of stock price. Different from previous studies, this paper emphasizes the fusion of individual features and multi-scale local features, and applies wavelet packet transform to denoise the data to improve the performance of the algorithm. The algorithm proposed in this paper is verified in the data set of CSI300 constituent stocks in the Chinese mainland. The performance of two benchmark indexes is 2% and 3% better than that of the common time series prediction baseline models.

References

[1]	张玉婧. 股市基本面分析与技术面分析结合方法探讨[J]. 商业文化(上半月), 2011(10): 201.
[2]	Fischer, T. and Krauss, C. (2017) Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions. European Journal of Operational Research, 270, 654-669. https://doi.org/10.1016/j.ejor.2017.11.054
[3]	Zhang, R., Yuan, Z. and Shao, X. (2018) A New Combined CNN-RNN Model for Sector Stock Price Analysis. 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC), Tokyo, 23-27 July 2018, 546-551. https://doi.org/10.1109/COMPSAC.2018.10292
[4]	赵红蕊, 薛雷. 基于LSTM-CNN-CBAM模型的股票预测研究[J]. 计算机工程与应用, 2021, 57(3): 203-207.
[5]	Xu, Y. and Cohen, S.B. (2018) Stock Movement Predic-tion from Tweets and Historical Prices. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Vol. 1: Long Papers, Melbourne, 15-20 July 2018, 1970-1979. https://doi.org/10.18653/v1/P18-1183
[6]	张梦吉, 杜婉钰, 郑楠. 引入新闻短文本的个股走势预测模型[J]. 数据分析与知识发现, 2019, 3(5): 11-18.
[7]	Kipf, T.N. and Welling, M. (2016) Semi-Supervised Classification with Graph Convolutional Networks. arXiv:1609.02907.
[8]	Chen, Y. and Wei, Z. and Huang, X. (2018) Incorporating Corporation Relationship via Graph Convolutional Neural Networks for Stock Price Prediction. Proceedings of the 27th ACM International Conference on Information and Knowledge Management, Torino, 22-26 October 2018, 1655-1658. https://doi.org/10.1145/3269206.3269269
[9]	Walczak, B. and Massart, D.L. (1997) Noise Suppression and Signal Compression Using the Wavelet Packet Transform. Chemometrics and Intelligent Laboratory Systems, 36, 81-94. https://doi.org/10.1016/S0169-7439(96)00077-9
[10]	Ramsey, J.B. (1999) The Contribution of Wavelets to the Analysis of Economic and Financial Data. Philosophical Transactions of the Royal Society A: Mathematical Physical & Engineering Sciences, 357, 1593-2606. https://doi.org/10.1098/rsta.1999.0450
[11]	郭晓霞, 杨慧中. 小波去噪中软硬阈值的一种改良折衷法[J]. 智能系统学报, 2008, 3(3): 222-225.
[12]	Bai, S., Kolter, J.Z. and Koltun, V. (2018) An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv preprint arXiv:1803.01271,.
[13]	Long, J., Shel-hamer, E. and Darrell, T. (2015) Fully Convolutional Networks for Semantic Segmentation. IEEE Transactions on Pat-tern Analysis and Machine Intelligence, 39, 640-651. https://doi.org/10.1109/TPAMI.2016.2572683
[14]	Van Den Oord, A., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A. and Kavukcuoglu, K. (2016) Wavenet: A Generative Model for Raw Audio. arXiv preprint arXiv:1609.03499.
[15]	Yu, F. and Koltun, V. (2016) Multi-Scale Context Aggregation by Dilated Convolutions. International Conference on Learning Representations 2016, San Juan, 2-4 May 2016, 1-13.
[16]	He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 770-778. https://doi.org/10.1109/CVPR.2016.90
[17]	Glorot, X., Bordes, A. and Bengio, Y. (2011) Deep Sparse Rectifier Neural Networks. Proceedings of the 14th International Conference on Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, 11-13 April 2011, 315-323.
[18]	Salimans, T. and Kingma, D.P. (2016) Weight Normalization: A Sim-ple Reparameterization to Accelerate Training of Deep Neural Networks. Advances in Neural Information Processing Systems, 29, 901-909.
[19]	Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. and Salakhutdinov, R. (2014) Dropout: A Simple Way to Prevent Neural Networks from Overfitting. Journal of Machine Learning Research, 15, 1929-1958.
[20]	何柏彬. 投资者情绪对我国股票价格波动的影响研究[D]: [硕士学位论文]. 杭州: 浙江大学, 2021.

Full-Text