Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LSR-IGRU: Stock Trend Prediction Based on Long Short-Term Relationships and Improved GRU

Peng Zhu, Yuante Li, Yifan Hu, Qinyuan Liu, Dawei Cheng, Yuqi Liang

TL;DR

LSR-IGRU addresses stock trend prediction by jointly modeling long-term and short-term stock relationships and integrating them into an improved GRU. Long-term links are built from secondary industry structure, while short-term links arise from overnight price-change cosine similarities, and both are embedded at every GRU step via a two-layer GAT. The approach yields superior predictive performance across CSI 300, CSI 500, S&P 500, and NASDAQ 100, and demonstrates practical value through deployment in an algorithmic trading platform with higher cumulative returns and controlled drawdown. These results highlight the importance of multi-scale relational information and tightly coupled temporal-relational modeling for robust financial forecasting.

Abstract

Stock price prediction is a challenging problem in the field of finance and receives widespread attention. In recent years, with the rapid development of technologies such as deep learning and graph neural networks, more research methods have begun to focus on exploring the interrelationships between stocks. However, existing methods mostly focus on the short-term dynamic relationships of stocks and directly integrating relationship information with temporal information. They often overlook the complex nonlinear dynamic characteristics and potential higher-order interaction relationships among stocks in the stock market. Therefore, we propose a stock price trend prediction model named LSR-IGRU in this paper, which is based on long short-term stock relationships and an improved GRU input. Firstly, we construct a long short-term relationship matrix between stocks, where secondary industry information is employed for the first time to capture long-term relationships of stocks, and overnight price information is utilized to establish short-term relationships. Next, we improve the inputs of the GRU model at each step, enabling the model to more effectively integrate temporal information and long short-term relationship information, thereby significantly improving the accuracy of predicting stock trend changes. Finally, through extensive experiments on multiple datasets from stock markets in China and the United States, we validate the superiority of the proposed LSR-IGRU model over the current state-of-the-art baseline models. We also apply the proposed model to the algorithmic trading system of a financial company, achieving significantly higher cumulative portfolio returns compared to other baseline methods. Our sources are released at https://github.com/ZP1481616577/Baselines_LSR-IGRU.

LSR-IGRU: Stock Trend Prediction Based on Long Short-Term Relationships and Improved GRU

TL;DR

LSR-IGRU addresses stock trend prediction by jointly modeling long-term and short-term stock relationships and integrating them into an improved GRU. Long-term links are built from secondary industry structure, while short-term links arise from overnight price-change cosine similarities, and both are embedded at every GRU step via a two-layer GAT. The approach yields superior predictive performance across CSI 300, CSI 500, S&P 500, and NASDAQ 100, and demonstrates practical value through deployment in an algorithmic trading platform with higher cumulative returns and controlled drawdown. These results highlight the importance of multi-scale relational information and tightly coupled temporal-relational modeling for robust financial forecasting.

Abstract

Stock price prediction is a challenging problem in the field of finance and receives widespread attention. In recent years, with the rapid development of technologies such as deep learning and graph neural networks, more research methods have begun to focus on exploring the interrelationships between stocks. However, existing methods mostly focus on the short-term dynamic relationships of stocks and directly integrating relationship information with temporal information. They often overlook the complex nonlinear dynamic characteristics and potential higher-order interaction relationships among stocks in the stock market. Therefore, we propose a stock price trend prediction model named LSR-IGRU in this paper, which is based on long short-term stock relationships and an improved GRU input. Firstly, we construct a long short-term relationship matrix between stocks, where secondary industry information is employed for the first time to capture long-term relationships of stocks, and overnight price information is utilized to establish short-term relationships. Next, we improve the inputs of the GRU model at each step, enabling the model to more effectively integrate temporal information and long short-term relationship information, thereby significantly improving the accuracy of predicting stock trend changes. Finally, through extensive experiments on multiple datasets from stock markets in China and the United States, we validate the superiority of the proposed LSR-IGRU model over the current state-of-the-art baseline models. We also apply the proposed model to the algorithmic trading system of a financial company, achieving significantly higher cumulative portfolio returns compared to other baseline methods. Our sources are released at https://github.com/ZP1481616577/Baselines_LSR-IGRU.
Paper Structure (25 sections, 5 equations, 3 figures, 2 tables)

This paper contains 25 sections, 5 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Traditional and Machine Learning Methods
  4. Deep and Reinforcement Learning Methods
  5. Graph Neural Networks and Latest Methods
  6. The Proposed Model LSR-IGRU
  7. Predefined
  8. Embedding of Long-term and Short-term Relationships of Stocks
  9. Constructing the long-term relationship matrix
  10. Constructing the short-term relationship matrix
  11. Using GAT to obtain representations of long-term and short-term relationships.
  12. Improved GRU Network
  13. Optimization Objectives
  14. Experiments
  15. Experimental Setttings
  16. ...and 10 more sections

Figures (3)

  • Figure 1: The architecture of the proposed model LST-IGRU. Part (a) outlines the generation of the long-short relationship representation model, which involves constructing long-term and short-term relationship matrices. These matrices are then inputted into respective GAT networks to derive relationship representations. Part (b) introduces enhancements to the GRU module, integrating both temporal and long-short relationship representations. Part (c) covers objective function optimization.
  • Figure 2: The results of parameter sensitivity experiments.
  • Figure 3: The performance of the strategy backtest.