Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models

Paper: arXiv 2304.07619

Authors: Alejandro Lopez-Lira, Yuehua Tang

Abstract

We document the capability of large language models (LLMs) like ChatGPT to predict stock market reactions from news headlines without direct financial training. Using post-knowledge-cutoff headlines, GPT-4 captures initial market responses, achieving approximately 90% portfolio-day hit rates for the non-tradable initial reaction. GPT-4 scores also significantly predict the subsequent drift, especially for small stocks and negative news. Forecasting ability generally increases with model size, suggesting that financial reasoning is an emerging capacity of complex LLMs. Strategy returns decline as LLM adoption rises, consistent with improved price efficiency. To rationalize these findings, we develop a theoretical model that incorporates LLM technology, information-processing capacity constraints, underreaction, and limits to arbitrage.

Complexity vs Empirical Score

• Math Complexity: 4.5/10
• Empirical Rigor: 8.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: This groundbreaking paper provides rigorous empirical evidence on the zero-shot capability of Large Language Models to forecast stock price movements from news headlines. While it includes a theoretical model to rationalize behavioral underreaction and limits to arbitrage, the primary strength lies in its extensive, well-validated empirical framework and high practical relevance.

Research Flowchart

  flowchart TD
    A[Research Goal: Evaluate LLMs' Ability<br>to Forecast Stock Returns from News] --> B[Input Data: Post-Cutoff News Headlines<br>and Stock Return Data]
    B --> C[Sentiment Scoring: GPT-4 & LLMs Rate<br>Headlines as Positive/Negative/Neutral]
    C --> D[Portfolio Construction: Form Long-Short<br>Strategies Based on LLM Scores]
    D --> E[Empirical Analysis: Measure Initial Market<br>Reactions & Post-Announcement Drift]
    E --> F[Theoretical Model: Rationalize Findings via<br>Information Constraints & Limits to Arbitrage]
    F --> G[Key Outcomes: ~90% Hit Rate for Initial Returns,<br>Predicts Drift in Small/Negative Stocks,<br>Returns Decline as LLM Adoption Rises]