基于生成式对抗神经网络的股票预测研究

doi:10.3778/j.issn.1002-8331.2112-0005

摘要/Abstract

摘要： 针对股票价格具有非线性、非平稳的特点，提出一种结合自注意力机制和残差网络的生成式对抗神经网络模型（SAR-GAN）。该模型的生成器（generator）由长短期记忆网络（LSTM）层、自注意力机制层、残差层等构建而成，用于生成所预测股票的价格；判别器（discriminator）用于鉴别生成的股票价格与真实的股票价格。为验证模型良好的泛化性，选取上证指数及不同股票市场的热点行业龙头股票进行预测实验。实验结果表明，与LSTM、GRU、CNN-LSTM、CNN-GRU等模型相比，SAR-GAN模型能不同程度地减少预测误差。

关键词: 股票预测, 生成式对抗神经网络, 自注意力机制

Abstract: Aiming at the nonlinear and non-stationary characteristics of stock price, this paper proposes a generative adversarial neural network model（SAR-GAN） combining self-attention mechanism and residual network. The generator of the model is composed of LSTM layer, self-attention mechanism layer and residual layer, which is used to generate the price of the predicted stock. Discriminator is used to distinguish the generated stock price from the real stock price. In order to verify the good generalization of the model, the Shanghai composite index and the leading stocks of hot industries in different stock markets are selected for prediction experiments. The experimental results show that compared with LSTM, GRU, CNN-LSTM and CNN-GRU models, the SAR-GAN model can reduce the prediction error in different degrees.

Key words: stock forecasting, generative adversarial networks（GAN）, self-attention mechanism

严冬梅, 李斌. 基于生成式对抗神经网络的股票预测研究[J]. 计算机工程与应用, 2022, 58(13): 185-194.

YAN Dongmei, LI Bin. Research on Stock Prediction Based on Generative Adversarial Networks[J]. Computer Engineering and Applications, 2022, 58(13): 185-194.

参考文献

[1] VOGL T P，MANGIS J K，RIGLER A K，et al.Accelerating the convergence of the back-propagation method[J].Biological Cybernetics，1988，59（4/5）：257-263.
[2] WHITE H.Economic prediction using neural networks：The case of IBM daily stock returns[C]//Proceedings of IEEE International Conference on Neural Networks，1988：451-458.
[3] 禹建丽，孙增圻，KROUMOV V，等.基于BP神经网络的股市建模与决策[J].系统工程理论与实践，2003，23（5）：15-19.
YU J L，SUN Z Q，KROUMOV V，et al.Stock market forecasting and decision-making based on neural network[J].Systems Engineering?Theory & Practice，2003，23（5）：15-19.
[4] 王莎.BP神经网络在股票预测中的应用研究[D].长沙：中南大学，2008.
WANG S.Application of BP neural network in stock forecasting[D].Changsha：Central South University，2008.
[5] 孙瑞奇.基于LSTM神经网络的美股股指价格趋势预测模型的研究[D].北京：首都经济贸易大学，2016.
SUN R Q.Research on the price trend prediction model of american stock index based on LSTM neural network[D].Beijing：Capital University of Economics and Business，2016.
[6] 邓凤欣，王洪良.LSTM神经网络在股票价格趋势预测中的应用——基于美港股票市场个股数据的研究[J].金融经济，2018，488（14）：98-100.
DENG F X，WANG H L.The application of LSTM neural network in stock price trend prediction based on the data of individual stocks in the US and Hong Kong stock markets[J].Financial Economics，2018，488（14）：98-100.
[7] 王东，王霄鹏，杨川东.一种基于主成分LSTM模型在股票预测中的研究[J].重庆理工大学学报（自然科学），2021，35（2）：282-288.
WANG D，WANG X P，YANG C D.A stucy of stock forecasting based on LSTM model of principal component analysis[J].Journal of Chongqing Institute of Technology（Natural Science），2021，35（2）：282-288.
[8] 邓飞燕，岑少琪，钟凤琪，等.基于LSTM神经网络的短期价格趋势预测[J].计算机系统应用，2021，30（4）：187-192.
DENG F Y，CEN S Q，ZHONG F Q，et al.Short-term price trend forecast based on LSTM neural network[J].Computer System & Applications，2021，30（4）：187-182.
[9] 郑树挺，徐菲菲.基于改进Self-Attention的股价趋势预测[J].计算机技术与发展，2021，31（3）：33-38.
ZHENG S T，XU F F.Research on stock price trend prediction based on Self-Attention model[J].Computer Technology and Development，2021，31（3）：33-38.
[10] 陈擎霄，刘传昌.基于自注意力的LSTM-CNN股票走势预测模型的研究[J/OL].中国科技论文在线[2021-02-24].http：//www.paper.edu.cn/releasepaper/content/202102-75.
CHEN Q X，LIU C C.Research on LSTM-CNN stock trend prediction model based on self-attention[J/OL].China Science and Technology Papers Online[2021-02-24].http：//www.paper.edu.cn/releasepaper/content/202102-75.
[11] 谷丽琼，吴运杰，逄金辉.基于Attention机制的GRU股票预测模型[J].系统工程，2020，38（5）：134-140.
GU L Q，WU Y J，PANG J H.An Attention-based mechanism in GRU model for stock predicting[J].Systems Engineering，2020，38（5）：134-140.
[12] 张倩玉，严冬梅，韩佳彤.结合深度学习和分解算法的股票价格预测研究[J].计算机工程与应用，2021，57（5）：56-64.
ZHANG Q Y，YAN D M，HAN J T.Research on stock price prediction combined with deep learning and decomposition algorithm[J].Computer Engineering and Applications，2021，57（5）：56-64.
[13] GOODFELLOW I J，POUGET-ABADIE J，MIRZA M，et al.Generative adversarial networks[J].Advances in Neural Information Processing Systems，2014，3：2672-2680.
[14] MIRZA M，OSINDERO S.Conditional generative adversarial nets[J].arXiv：1411.1784，2014.
[15] RADFORD A，METZ L，CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[J].arXiv：1511.06434，2015.
[16] KOOCHALI A，SCHICHTEL P，AHMED S，et al.Probabilistic forecasting of sensory data with generative adversarial networks-ForGAN[J].IEEE Access，2019，7：63868-63880.
[17] BAHDANAU D，CHO K，BENGIO Y.Neural machine translation by jointly learning to align and translate[J].arXiv：1409.0473，2014.
[18] FU J，ZHENG H，TAO M.Look closer to see better：Recurrent attention convolutional neural network for fine-grained image recognition[C]//Proceedings of IEEE Conference on Computer Vision & Pattern Recognition，2017.