Signal or Noise in Multi-Agent LLM-based Stock Recommendations?

2026-04-19

AI Summary

This study investigates whether stock recommendation signals generated by a multi-agent large language model (LLM) system can deliver statistically significant excess returns and identifies their informational origins. We construct a framework comprising four specialized agents—news, fundamentals, dynamics, and macro—and synthesize monthly recommendations for backtesting in a forward-looking, bias-free manner. Our findings reveal that agent contributions adaptively shift with market regimes, and the resulting signal serves as an effective stock-selection filter, capturing alpha beyond traditional factor models. Applied to the S&P 500 universe, the strategy achieves an average monthly excess return of +1.03% (annualized +25.2%, p = 0.003) and an information ratio of 0.489 (p = 0.024), significantly outperforming passive benchmarks.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Agent Communication
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

We present the first portfolio-level validation of MarketSenseAI, a deployed multi-agent LLM equity system. All signals are generated live at each observation date, eliminating look-ahead bias. The system routes four specialist agents (News, Fundamentals, Dynamics, and Macro) through a synthesis agent that issues a monthly equity thesis and recommendation for each stock in its coverage universe, and we ask two questions: do its buy recommendations add value over both passive benchmarks and random selection, and what does the internal agent structure reveal about the source of the edge? On the S&P 500 cohort (19 months) the strong-buy equal-weight portfolio earns +2.18%/month against a passive equal-weight benchmark of +1.15% (approximating RSP), a +25.2% compound excess, and ranks at the 99.7th percentile of 10,000 Monte Carlo portfolios (p=0.003). The S&P 100 cohort (35 months) delivers a +30.5% compound excess over EQWL with consistent direction but formal significance not reached, limited by the small average selection of ~10 stocks per month. Non-negative least-squares projection of thesis embeddings onto agent embeddings reveals an adaptive-integration mechanism. Agent contributions rotate with market regime (Fundamentals leads on S&P 500, Macro on S&P 100, Dynamics acts as an episodic momentum signal) and this agent rotation moves in lockstep with both the sector composition of strong-buy selections and identifiable macro-calendar events, three independent views of the same underlying adaptation. The recommendation's cross-sectional Information Coefficient is statistically significant on S&P 500 (ICIR=+0.489, p=0.024). These results suggest that multi-agent LLM equity systems can identify sources of alpha beyond what classical factor models capture, and that the buy signal functions as an effective universe-filter that can sit upstream of any portfolio-construction process.

Problem

• multi-agent LLM
• stock recommendation
• alpha generation
• portfolio validation
• market regime

Innovation

• multi-agent LLM
• adaptive integration
• look-ahead bias elimination
• portfolio-level validation
• Information Coefficient