Sequential Graph Attention Learning for Predicting Dynamic Stock Trends (Student Abstract)

TL;DR

This paper introduces a novel sequential graph attention model that leverages temporal and relational data to improve stock trend predictions across various industries, outperforming existing methods on Taiwan Stock datasets.

Contribution

The study proposes the GAT-AGNN framework, integrating sequential graph structures with attention mechanisms for enhanced dynamic stock trend prediction.

Findings

Outperforms state-of-the-art methods in stock trend prediction

Effective across multiple industries on Taiwan Stock datasets

Demonstrates the importance of combining temporal and relational information

Abstract

The stock market is characterized by a complex relationship between companies and the market. This study combines a sequential graph structure with attention mechanisms to learn global and local information within temporal time. Specifically, our proposed "GAT-AGNN" module compares model performance across multiple industries as well as within single industries. The results show that the proposed framework outperforms the state-of-the-art methods in predicting stock trends across multiple industries on Taiwan Stock datasets.