Inductive Representation Learning on Dynamic Stock Co-Movement Graphs for Stock Predictions

doi:10.1287/ijoc.2022.1172

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 互联网大数据与安全信息学研究中心

	Inductive Representation Learning on Dynamic Stock Co-Movement Graphs for Stock Predictions
	Tian, Hu 3,4; Zheng, Xiaolong 3,4; Zhao, Kang 2; Liu, Maggie Wenjing 1; Zeng, Daniel Dajun 3,4
刊名	INFORMS JOURNAL ON COMPUTING
	2022-03-01
页码	19
关键词	graph representation learning deep learning predictive models business intelligence
ISSN号	1091-9856
DOI	10.1287/ijoc.2022.1172
通讯作者	Zheng, Xiaolong(xiaolong.zheng@ia.ac.cn)
英文摘要	Co-movement among individual firms' stock prices can reflect complex inter firm relationships. This paper proposes a novel method to leverage such relationships for stock price predictions by adopting inductive graph representation learning on dynamic stock graphs constructed based on historical stock price co-movement. To learn node representations from such dynamic graphs for better stock predictions, we propose the hybrid-attention dynamic graph neural network, an inductive graph representation learning method. We also extended mini-batch gradient descent to inductive representation learning on dynamic stock graphs so that the model can update parameters over mini batch stock graphs with higher training efficiency. Extensive experiments on stocks from different markets and trading simulations demonstrate that the proposed method signifi-cantly improves stock predictions. The proposed method can have important implications for the management of financial portfolios and investment risk. Summary of Contribution: Accurate predictions of stock prices have important implications for financial decisions. In today's economy, individual firms are increasingly connected via different types of relationships. As a result, firms' stock prices often feature synchronous co-movement patterns. This paper represents the first effort to leverage such phenomena to construct dynamic stock graphs for stock predictions. We develop hybridattention dynamic graph neural network (HAD-GNN), an inductive graph representation learning framework for dynamic stock graphs to incorporate temporal and graph attention mechanisms. To improve the learning efficiency of HAD-GNN, we also extend the minibatch gradient descent to inductive representation learning on such dynamic graphs and adopt a t-batch training mechanism (t-BTM). We demonstrate the effectiveness of our new approach via experiments based on real-world data and simulations.
资助项目	Ministry of Science and Technology of China[2020AAA0108401] ; Natural Science Foundation of China[71621002] ; Natural Science Foundation of China[71472175] ; Natural Science Foundation of China[71602184] ; Natural Science Foundation of China[71991462] ; Natural Science Foundation of China[71825007] ; Ministry ofHealth of China[2017ZX10303401-002] ; Strategic Priority Research Programof Chinese Academy of Sciences[XDA27030100]
WOS关键词	BEHAVIOR ; RETURNS ; NETWORK
WOS研究方向	Computer Science ; Operations Research & Management Science
语种	英语
出版者	INFORMS
WOS记录号	WOS:000803709800001
资助机构	Ministry of Science and Technology of China ; Natural Science Foundation of China ; Ministry ofHealth of China ; Strategic Priority Research Programof Chinese Academy of Sciences
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/49521]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_互联网大数据与安全信息学研究中心
通讯作者	Zheng, Xiaolong
作者单位	1.Tsinghua Univ, Sch Econ & Management, Beijing 100084, Peoples R China 2.Univ Iowa, Tippie Coll Business, Dept Business Analyt, Iowa City, IA 52242 USA 3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100190, Peoples R China 4.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Tian, Hu,Zheng, Xiaolong,Zhao, Kang,et al. Inductive Representation Learning on Dynamic Stock Co-Movement Graphs for Stock Predictions[J]. INFORMS JOURNAL ON COMPUTING,2022:19.
APA	Tian, Hu,Zheng, Xiaolong,Zhao, Kang,Liu, Maggie Wenjing,&Zeng, Daniel Dajun.(2022).Inductive Representation Learning on Dynamic Stock Co-Movement Graphs for Stock Predictions.INFORMS JOURNAL ON COMPUTING,19.
MLA	Tian, Hu,et al."Inductive Representation Learning on Dynamic Stock Co-Movement Graphs for Stock Predictions".INFORMS JOURNAL ON COMPUTING (2022):19.