Stock trend prediction method based on temporal hypergraph convolutional neural network

doi:10.11772/j.issn.1001-9081.2021050748

Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (3): 797-803.DOI: 10.11772/j.issn.1001-9081.2021050748

• 2021 CCF Conference on Artificial Intelligence (CCFAI 2021) • Previous Articles

Stock trend prediction method based on temporal hypergraph convolutional neural network

Xiaojie LI¹, Chaoran CUI¹(), Guangle SONG², Yaxi SU¹, Tianze WU³, Chunyun ZHANG¹

^1.School of Computer Science and Technology，Shandong University of Finance and Economics，Jinan Shandong 250014，China
^2.Shandong Association for Artificial Intelligence，Jinan Shandong 250101，China
^3.School of Computer Science and Technology，Qilu University of Technology （Shandong Academy of Sciences），Jinan Shandong 250353，China

Received:2021-05-11 Revised:2021-07-16 Accepted:2021-07-21 Online:2021-11-09 Published:2022-03-10
Contact: Chaoran CUI
About author:LI Xiaojie， born in 1998， M. S. candidate. Her research interests include deep learning， data mining.
SONG Guangle， born in 1992， M. S. His research interests include computer vision， machine learning， artificial intelligence.
SU Yaxi， born in 1997， M. S. candidate. Her research interests include deep learning， data mining.
WU Tianze， born in 1991， M. S.， lecturer. His research interests include data mining.
ZHANG Chunyun， born in 1986， Ph. D.， associate professor. Her research interests include natural language processing， machine learning.
Supported by:
National Natural Science Foundation of China(62077033);National Key Research and Development Program of China(2018YFC0830100);Shandong Provincial Natural Science Foundation Key Project(ZR2020KF015);Fostering Project of Dominant Discipline and Talent Team of Shandong Province Higher Education Institutions

基于时序超图卷积神经网络的股票趋势预测方法

李晓杰¹, 崔超然¹(), 宋广乐², 苏雅茜¹, 吴天泽³, 张春云¹

^1.山东财经大学计算机科学与技术学院, 济南 250014
^2.山东省人工智能学会, 济南 250101
^3.齐鲁工业大学(山东省科学院) 计算机科学与技术学院, 济南 250353

通讯作者: 崔超然
作者简介:李晓杰（1998—），女，山东聊城人，硕士研究生，主要研究方向：深度学习、数据挖掘
宋广乐（1992—），男，山东德州人，硕士，主要研究方向：计算机视觉、机器学习、人工智能
苏雅茜（1997—），女，山东济南人，硕士研究生，主要研究方向：深度学习、数据挖掘
吴天泽（1991—），男，山东济南人，讲师，硕士，主要研究方向：数据挖掘
张春云（1986—），女，山东聊城人，副教授，博士，主要研究方向：自然语言处理、机器学习。
基金资助:
国家自然科学基金资助项目(62077033);国家重点研发计划项目(2018YFC0830100);山东省自然科学基金重点项目(ZR2020KF015);山东省高等学校优势学科人才团队培育计划项目

Abstract

Abstract:

Traditional stock prediction methods are mostly based on time-series models， which ignore the complex relations among stocks， and the relations often exceed pairwise connections， such as stocks in the same industry or multiple stocks held by the same fund. To solve this problem， a stock trend prediction method based on temporal HyperGraph Convolutional neural Network （HGCN） was proposed， and a hypergraph model based on financial investment facts was constructed to fit multiple relations among stocks. The model was composed of two major components： Gated Recurrent Unit （GRU） network and HGCN. GRU network was used for performing time-series modeling on historical data to capture long-term dependencies. HGCN was used to model high-order relations among stocks to learn intrinsic relation attributes， and introduce the multiple relation information among stocks into traditional time-series modeling for end-to-end trend prediction. Experiments on real dataset of China A-share market show that compared with existing stock prediction methods， the proposed model improves prediction performance， e.g. compared with the GRU network， the proposed model achieves the relative increases in ACC and F1_score of 9.74% and 8.13%， respectively， and is more stable. In addition， the simulation back-testing results show that the trading strategy based on the proposed model is more profitable， with an annual return of 11.30%， which is 5 percentage points higher than that of Long Short-Term Memory （LSTM） network.

Key words: stock trend prediction, time series modeling, Gated Recurrent Unit (GRU), high-order relation, HyperGraph Convolutional neural Network (HGCN)

摘要：

传统的股票预测方法大多基于时间序列模型，忽视了股票之间复杂的关系，并且该关系往往超出成对连接，例如同行业板块内股票或者基金持仓多支股票。针对该问题，提出一种基于时序超图卷积神经网络（HGCN）的股价走势预测方法，根据金融投资事实构造超图模型以拟合股票之间的多元关系，该模型包括两大组件：门控循环单元（GRU）网络和超图卷积神经网络。GRU网络对历史数据进行时间序列建模，捕捉长期依赖关系；HGCN建模股票间的高阶关系以学习内在关系属性，从而将股票间多元关系信息引入到传统的时序建模中，进行端到端的趋势预测。在中国A股市场真实数据集上的实验结果表明，相较于已有的股票预测方法，所提模型预测性能有所提升；如与GRU网络相比，所提模型在ACC和F1_score上的相对增幅分别为9.74%和8.13%，且更具有稳定性。此外，模拟回测结果显示，基于该模型的交易策略更具获利能力，年回报率达到11.30%，与长短期记忆（LSTM）网络相比提高了5个百分点。

关键词: 股票趋势预测, 时间序列建模, 门控循环单元, 高阶关系, 超图卷积神经网络

CLC Number:

TP181

Xiaojie LI, Chaoran CUI, Guangle SONG, Yaxi SU, Tianze WU, Chunyun ZHANG. Stock trend prediction method based on temporal hypergraph convolutional neural network[J]. Journal of Computer Applications, 2022, 42(3): 797-803.

李晓杰, 崔超然, 宋广乐, 苏雅茜, 吴天泽, 张春云. 基于时序超图卷积神经网络的股票趋势预测方法[J]. 《计算机应用》唯一官方网站, 2022, 42(3): 797-803.

Figures/Tables 9

References 22

1	HINTON G E， OSINDERO S， TEH Y W. A fast learning algorithm for deep belief nets ［J］. Neural Computation， 2006， 18（7）： 1527-1554. 10.1162/neco.2006.18.7.1527
2	SCHUSTER M， KULDIP K P. Bidirectional recurrent neural networks ［J］. IEEE Transactions on Signal Processing， 1997， 45（11）： 2673-2681. 10.1109/78.650093
3	HOCHREITER S， SCHMIDHUBER J. Long short-term memory ［J］. Neural Computation， 1997， 9（8）： 1735-1780. 10.1162/neco.1997.9.8.1735
4	CHO K， MERRIENBOER B V， GULCEHRE C， et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation ［C］// Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Stroudsburg， PA： ACL， 2014： 1724-1734. 10.3115/v1/d14-1179
5	NIEPERT M， AHMED M， KUTZKOV K. Learning convolutional neural networks for graphs ［C］// Proceedings of the 33rd International Conference on Machine Learning. New York： ACM， 2016： 2014-2023.
6	PREOZZI B， ALRFOU R， SKIENA S. DeepWalk： online learning of social representations ［C］// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2014： 701-710.
7	CHEN Y， WEI Z， HUANG X. Incorporating corporation relationship via graph convolutional neural networks for stock price prediction ［C］// Proceedings of the 27th ACM International Conference on Information and Knowledge Management. New York： ACM， 2018： 1655-1658. 10.1145/3269206.3269269
8	FENG F， HE X， WANG X， et al. Temporal relational ranking for stock prediction ［EB/OL］.［2020-06-20］.. 10.1145/3309547
9	FENG Y， YOU H， ZHANG Z， et al. Hypergraph neural networks ［C］// Proceedings of the 33rd Conference on Association for the Advancement of Artificial Intelligence. Menlo Park， CA： AAAI， 2019： 3558-3565. 10.1609/aaai.v33i01.33013558
10	杨种学. 基于回归技术商品销售趋势预测模型的实现［J］. 保山师专学报， 2009， 28（5）： 64-67. 10.3969/j.issn.1674-9340.2009.05.021
	YANG Z X. Implementation of transaction trend prediction model based on regression analysis ［J］. Journal of Baoshan Teachers’ College， 2009， 28（5）： 64-67. 10.3969/j.issn.1674-9340.2009.05.021
11	OKASHA M K. Using support vector machines in financial time series forecasting ［J］. International Journal of Statistics and Applications， 2014， 4（1）： 28-39.
12	邓晶，李路. 参数优化随机森林在股票预测中的应用［J］. 软件， 2020， 41（1）： 178-182. 10.3969/j.issn.1003-6970.2020.01.039
	DENG J， LI L. Application of parameter optimization random forest in stock forecasting ［J］. Software， 2020， 41（1）： 178-182. 10.3969/j.issn.1003-6970.2020.01.039
13	刘建伟，王园方，罗雄麟. 深度记忆网络研究进展［J］. 计算机学报， 2020， 43（3）： 2-52.
	LIU J W， WANG Y F， LUO X L. Research progress of deep memory networks ［J］. Chinese Journal of Computer Science， 2020， 43（3）： 2-52.
14	AKITA R， YOSHIHARA A， MATSUBARA T， et al. Deep learning for stock prediction using numerical and textual information ［C］// Proceedings of the 15th International Conference on Computer and Information Science. Piscataway： IEEE， 2016： 1-6. 10.1109/icis.2016.7550882
15	ZHANG L， AGGARWAL C， QI G. Stock price prediction via discovering multi-frequency trading patterns ［C］// Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2017： 2141-2149. 10.1145/3097983.3098117
16	黎镭，陈蔼祥，李伟书，等. GRU递归神经网络对股票收盘价的预测研究［J］. 计算机与现代化， 2018（11）： 103-108. 10.3969/j.issn.1006-2475.2018.11.018
	LI L， CHEN A X， LI W S， et al. Research on the prediction of stock closing price by GRU recursive neural network ［J］. Computer and Modernization， 2018（11）： 103-108. 10.3969/j.issn.1006-2475.2018.11.018
17	WANG M， XUE L， WU X. Visual classification by l1-hypergraph modeling ［J］. IEEE Transactions on Knowledge and Data Engineering， 2015， 27（9）：2564-2574.
18	HUANG Y， LIU Q， METAXAS D. Video object segmentation by hypergraph cut ［C］// Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway， NJ： IEEE， 2009： 1738-1745. 10.1109/cvpr.2009.5206795
19	CHEN F， GAO Y， CAO D， et al. Multimodal hypergraph learning for microblog sentiment prediction ［C］// Proceedings of the 2015 IEEE International Conference on Multimedia and Expo. Piscataway， NJ： IEEE， 2015： 1-6. 10.1109/icme.2015.7177477
20	JI R， CHEN F， CAO L， et al. Cross-modality microblog sentiment prediction via bi-layer multimodal hypergraph learning ［J］. IEEE Transactions on Multimedia， 2019， 21（4）： 1062-1075. 10.1109/tmm.2018.2867718
21	BAI S， ZHANG F， TORR P H. Hypergraph convolution and hypergraph attention ［J］. Pattern Recognition， 2021， 110： 107637. 10.1016/j.patcog.2020.107637
22	FENG F， CHEN H， HE X， et al. Enhancing stock movement prediction with adversarial training ［C］// Proceedings of the 28th International Joint Conference on Artificial Intelligence. Menlo Park， CA： AAAI， 2019： 5843-5849. 10.24963/ijcai.2019/810

数据集	交易天数	上涨比率	下跌比率	持平比率
训练集	1 021	0.385	0.375	0.240
验证集	340	0.379	0.372	0.249
测试集	331	0.363	0.360	0.277

数据集	交易天数	上涨比率	下跌比率	持平比率
训练集	1 021	0.385	0.375	0.240
验证集	340	0.379	0.372	0.249
测试集	331	0.363	0.360	0.277

股票代码	股票名称	所属行业	股票代码	股票名称	所属行业
000001.SZ	平安银行	银行	000089.SZ	深圳机场	机场
000011.SZ	深物业A	区域地产	000338.SZ	潍柴动力	汽车配件
000012.SZ	南玻A	玻璃	000417.SZ	合肥百货	百货
000014.SZ	沙河股份	全国地产	000419.SZ	通程控股	百货
000021.SZ	深科技	IT设备	000423.SZ	东阿阿胶	中成药
000030.SZ	富奥股份	汽车配件	000501.SZ	鄂武商A	百货
000049.SZ	德赛电池	电气设备	000507.SZ	珠海港	港口
000088.SZ	盐田港	港口

股票代码	股票名称	所属行业	股票代码	股票名称	所属行业
000001.SZ	平安银行	银行	000089.SZ	深圳机场	机场
000011.SZ	深物业A	区域地产	000338.SZ	潍柴动力	汽车配件
000012.SZ	南玻A	玻璃	000417.SZ	合肥百货	百货
000014.SZ	沙河股份	全国地产	000419.SZ	通程控股	百货
000021.SZ	深科技	IT设备	000423.SZ	东阿阿胶	中成药
000030.SZ	富奥股份	汽车配件	000501.SZ	鄂武商A	百货
000049.SZ	德赛电池	电气设备	000507.SZ	珠海港	港口
000088.SZ	盐田港	港口

截止日期	股票代码										所含股票数
2013-03-31	000639.SZ	000830.SZ	000999.SZ	002022.SZ	002570.SZ	002664.SZ	300032.SZ	600285.SH	600527.SH	600571.SH	10
2013-06-30	000639.SZ	000830.SZ	000999.SZ	002022.SZ	002570.SZ	300032.SZ	600486.SH	600527.SH	…	600887.SH	33
2013-09-30	000538.SZ	000651.SZ	000661.SZ	002236.SZ	002241.SZ	002415.SZ	300058.SZ	300215.SZ	300228.SZ	601633.SH	10
2013-12-31	000538.SZ	000651.SZ	002236.SZ	002241.SZ	002353.SZ	300215.SZ	300228.SZ	600309.SH	…	601628.SH	22

Stock trend prediction method based on temporal hypergraph convolutional neural network

基于时序超图卷积神经网络的股票趋势预测方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 22

Related Articles 10

Recommended Articles

Metrics

方法	Accuracy	Precision	Recall	F1_score
GRU+GCN_industry	37.232±0.111	35.003±2.962	34.255±3.150	34.611±2.925
GRU+GCN_fund	37.076±0.259	34.123±1.197	34.172±0.412	34.144±0.792
GRU+GCN_hybrid	37.570±0.223	38.497±1.696	34.445±0.567	36.352±1.035
GRU+HGCN_industry	38.407±0.442	38.361±0.060	34.815±0.170	36.502±0.114
GRU+HGCN_fund	38.292±0.448	37.482±1.102	34.805±0.915	36.090±0.866
GRU+HGCN_hybrid	38.852±0.361	39.593±1.256	35.652±0.365	37.514±0.689

模型	Accuracy	Precision	Recall	F1_score
LSTM	35.187±0.161	35.963±8.260	34.444±0.240	34.868±4.375
GRU	35.402±1.239	36.154±8.529	33.975±0.452	34.694±4.398
SFM	33.314±0.201	26.498±6.912	33.336±0.002	29.213±4.058
RSR	37.560±1.010	34.470±9.584	35.502±0.717	34.547±5.708
GRU+HGCN_hybrid	38.852±0.361	39.593±1.256	35.652±0.365	37.514±0.689

方法	R/%	AR/%	SR	MDD/%
Buy-and-hold	2.48	2.16	0.027	20.22
LSTM	6.84	5.96	0.271	13.92
GRU	6.43	5.58	0.240	14.48
SFM	5.78	5.02	0.523	5.38
RSR	3.20	2.78	0.067	16.72
GRU+HGCN_hybrid	13.07	11.30	0.550	12.40

[1]	Donglin MA, Sizhou MA, Weijie WANG. Multi-site temperature prediction model based on graph convolutional network and gated recurrent unit [J]. Journal of Computer Applications, 2022, 42(1): 287-293.
[2]	LIU Zichen, LI Xiaojuan, WEI Wei. Automatic patent price evaluation based on recurrent neural network [J]. Journal of Computer Applications, 2021, 41(9): 2532-2538.
[3]	Huibo LI, Yunxiao ZHAO, Liang BAI. Dynamic graph representation learning method based on deep neural network and gated recurrent unit [J]. Journal of Computer Applications, 2021, 41(12): 3432-3437.
[4]	MENG Xinyu, WANG Ruihan, ZHANG Xiping, WANG Mingjie, QIU Gang, WANG Zhengxia. Ultra-short-term wind power prediction based on empirical mode decomposition and multi-branch neural network [J]. Journal of Computer Applications, 2021, 41(1): 237-242.
[5]	GAO Shile, WANG Ying, LI Hailin, WAN Xiaoji. Prediction method on financial time series data based on matrix profile [J]. Journal of Computer Applications, 2021, 41(1): 199-207.
[6]	YANG Yunlong, SUN Jianqiang, SONG Guochao. Text sentiment analysis based on gated recurrent unit and capsule features [J]. Journal of Computer Applications, 2020, 40(9): 2531-2535.
[7]	CHEN Xiukai, LU Zhihua, ZHOU Yu. Speech separation algorithm based on convolutional encoder decoder and gated recurrent unit [J]. Journal of Computer Applications, 2020, 40(7): 2137-2141.
[8]	HUANG Weijian, LI Danyang, HUANG Yuan. Spatio-temporal hybrid prediction model for air quality [J]. Journal of Computer Applications, 2020, 40(11): 3385-3392.
[9]	ZHANG Dezheng, WENG Liguo, XIA Min, CAO Hui. Video frame prediction based on deep convolutional long short-term memory neural network [J]. Journal of Computer Applications, 2019, 39(6): 1657-1662.
[10]	ZHOU Xiangyu, CHENG Yong, WANG Jun. Agricultural greenhouse temperature prediction method based on improved deep belief network [J]. Journal of Computer Applications, 2019, 39(4): 1053-1058.