Using ChatGPT News Signals to Improve Equity Momentum Portfolios

Summary

This study tests whether a large language model can use company news to refine a cross-sectional momentum strategy. It combines daily returns for S&P 500 constituents with high-frequency news, then prompts ChatGPT to judge whether recent news supports continuation of a stock's past performance. The resulting scores influence which stocks enter the portfolio and how positions are weighted.

The authors report that the enhanced strategy beats a long-only momentum benchmark on risk-adjusted measures in both in-sample and out-of-sample periods, including a period after the model's training-data cutoff. They also describe the results as robust to transaction costs, prompt choices, and portfolio constraints, with stronger gains in concentrated portfolios. The evidence concerns a particular prompt-based model, equity universe, and strategy design; the summary does not provide effect sizes or enough detail to assess implementation assumptions or generalization to other markets and models.

Key ideas

• The approach uses a language model to interpret firm news in the context of recent stock momentum.
• Model scores affect both stock selection and portfolio weights.
• The reported results exceed a long-only momentum benchmark on Sharpe and Sortino measures.
• The authors report robustness to costs, prompt design, and portfolio constraints.
• Reported gains are strongest in concentrated portfolios, so results may depend on portfolio construction.

Tags

# ChatGPT in Systematic Investing -- Enhancing Risk-Adjusted Returns with LLMs


# ChatGPT in Systematic Investing -- Enhancing Risk-Adjusted Returns with LLMs









This paper investigates whether large language models (LLMs) can improve cross-sectional momentum strategies by extracting predictive signals from firm-specific news. We combine daily U.S. equity returns for S&P 500 constituents with high-frequency news data and use prompt-engineered queries to ChatGPT that inform the model when a stock is about to enter a momentum portfolio. The LLM evaluates whether recent news supports a continuation of past returns, producing scores that condition both stock selection and portfolio weights. An LLM-enhanced momentum strategy outperforms a standard long-only momentum benchmark, delivering higher Sharpe and Sortino ratios both in-sample and in a truly out-of-sample period after the model's pre-training cut-off. These gains are robust to transaction costs, prompt design, and portfolio constraints, and are strongest for concentrated, high-conviction portfolios. The results suggest that LLMs can serve as effective real-time interpreters of financial news, adding incremental value to established factor-based investment strategies.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.