ChatGPT Informed Graph Neural Network for Stock Movement Prediction
Zihan Chen, Lei Nico Zheng, Cheng Lu, Jialu Yuan, Di Zhu
TL;DR
This work addresses stock movement prediction under dynamic inter-stock dependencies by leveraging ChatGPT to infer evolving networks from daily financial news and feed them into a Graph Neural Network. The authors build a three-component framework that uses ChatGPT-based network inference to generate a time-varying graph, a GNN to produce company embeddings, and sequential models to forecast next-day movements. Experiments on Dow 30 data show consistent improvements over strong baselines in F1 metrics and portfolio performance, including higher cumulative returns and lower risk. The work demonstrates the potential of integrating LLM-driven network inferences with graph-based predictive models for practical finance applications.
Abstract
ChatGPT has demonstrated remarkable capabilities across various natural language processing (NLP) tasks. However, its potential for inferring dynamic network structures from temporal textual data, specifically financial news, remains an unexplored frontier. In this research, we introduce a novel framework that leverages ChatGPT's graph inference capabilities to enhance Graph Neural Networks (GNN). Our framework adeptly extracts evolving network structures from textual data, and incorporates these networks into graph neural networks for subsequent predictive tasks. The experimental results from stock movement forecasting indicate our model has consistently outperformed the state-of-the-art Deep Learning-based benchmarks. Furthermore, the portfolios constructed based on our model's outputs demonstrate higher annualized cumulative returns, alongside reduced volatility and maximum drawdown. This superior performance highlights the potential of ChatGPT for text-based network inferences and underscores its promising implications for the financial sector.