The Predictive Edge

CHAPTER 6Can ChatGPT Forecast Stock Price Movements?

Predicting stock market trends has been a long‐standing quest in finance. Investors and analysts have traditionally relied on various methods to predict stock movements, including fundamental analysis, technical analysis, and quantitative models. However, large language models (LLMs) like ChatGPT have begun to transform this landscape dramatically.

Historically, financial markets have been analyzed using structured data, such as financial ratios and market indices. As technology progressed, the focus shifted to unstructured data, including news articles, social media posts, and financial reports, to understand market sentiments. The advent of machine learning models provided tools to process and analyze these vast, complex datasets more efficiently and accurately than ever before.

This chapter examines an empirical study (coauthored with Yuehua Tang) that explores whether ChatGPT can predict stock returns using news headlines (Lopez‐Lira and Tang 2023). The study is motivated by a significant shift in the AI landscape, marked by the emergence of advanced LLMs capable of understanding and generating human‐like text. ChatGPT, a GPT (Generative Pretrained Transformer) model variant, represents a leap forward. Its ability to understand context, nuance, and the subtleties of human language makes it an intriguing tool for financial analysis, particularly in interpreting news and other textual data that influence stock markets. ...