HomeArticlesServicePriceAbout

A recent study, Can ChatGPT Forecast Stock Price Movements?, examines whether large language models (LLMs) like ChatGPT can predict stock price movements based on news headlines. The findings reveal that ChatGPT-generated sentiment scores significantly forecast next-day stock returns in out-of-sample testing. The predictive power is stronger for small-cap stocks and those with negative news, suggesting that LLMs are particularly effective at interpreting complex or underreported financial information. The study also proposes that widespread LLM adoption could improve market efficiency, reducing arbitrage opportunities over time.

Methodology

Data Collection:

• Gathered U.S. stock market news headlines and daily returns from October 2021 to December 2023.

ChatGPT Sentiment Analysis:

• Used ChatGPT to classify headlines as positive, negative, or neutral.

Portfolio Construction:

• Built long-short portfolios based on ChatGPT sentiment scores.

Regression Analysis:

• Controlled for factors such as news complexity, company size, and industry effects.

Out-of-Sample Predictability:

• Tested whether ChatGPT scores could predict next-day stock returns, assessing their economic value in real-world trading.

Key Findings

1. ChatGPT’s Sentiment Scores Predict Stock Returns

• ChatGPT’s sentiment scores are statistically significant predictors of next-day returns.

• Stocks with positive sentiment scores tend to rise, while those with negative sentiment decline.

2. Stronger Predictability for Small-Cap Stocks & Negative News

• Small-cap stocks exhibit stronger predictability, likely due to less analyst coverage and higher information asymmetry.

• Negative news drives sharper price movements, suggesting that ChatGPT effectively captures downside risk.

3. Model Complexity Matters

• Larger LLMs show better predictive accuracy, as they can decode complex financial narratives more effectively.

4. LLM Adoption May Improve Market Efficiency

• As more investors use LLM-based analytics, information asymmetries could diminish, making it harder to exploit sentiment-based predictions.

• This suggests that ChatGPT’s edge in predicting stock returns may erode over time as LLMs become standard tools in financial markets.

The study confirms that ChatGPT is a valuable tool for analyzing news sentiment and predicting stock price movements, particularly in small-cap and negative-news scenarios. However, as LLMs become widely adopted, their ability to generate alpha may decline, reinforcing their role in enhancing market efficiency rather than sustaining arbitrage opportunities.

[Compliance Note]

• All posts by Sellsmart are for informational purposes only. Final investment decisions should be made with careful judgment and at the investor’s own risk.

• The content of this post may be inaccurate, and any profits or losses resulting from trades are solely the responsibility of the investor.

• Core16 may hold positions in the stocks mentioned in this post and may buy or sell them at any time.