Agent communication influences stock holding similarity in trading swarms
Test-Time Communication in Trading Agent Swarms
Multiagent SystemsArtificial IntelligenceComputation and Language
Summary
Trading agents usually act on their own, making decisions that are judged by slow and noisy market results. The authors studied how agents communicating while trading affected their choices and returns over several days. They found that some groups of communicating agents tended to hold more of the same stocks than those working alone. Seeing messages from others made agents more likely to reuse phrases, but did not immediately change their trades. This suggests that group communication leads to more coordinated selling and buying over time, but isolated agents also sometimes chose the same stocks, so measuring herding needs to compare with solo behavior.
What this means in practice
- •For quantitative trading teams: Incorporate peer message sharing simulations into algorithm design to better predict group stock holding patterns and potential coordinated trading behavior.
- •For financial data platform developers: Design market analysis tools that track communication effects on trade convergence to improve detection of excessive herding beyond baseline market behavior.
Authors
Avid Fayaz, Harry Zhang, Håvard Alstadheim, Erik Nordlöf, Vilhelm Karlin, André Silva
Abstract
Communication between agents has mostly been studied on tasks with a shared objective and fast, clear feedback. Trading has neither: each agent manages its own account, with decisions being scored by markets late and noisily. A trading swarm is thus a population of independent decision-makers rather than a team, for which convergence is a risk and not the goal. We ran 200 agents from four model families in groups of ten under five communication regimes, from isolation to message boards and shared notes, over seven US trading days (1 to 9 October 2026), logging everything each agent read before every decision. In two of the four families, communicating groups came to hold far more of the same stocks than the isolated group. A group's return was largely the return on what its members held in common. Seeing a post from a groupmate made an agent 7 percentage points more likely to reuse an eight-word phrase from it. Agents that started too close together to see each other's posts shared such a phrase in 14% of cases, the rate that common inputs produce. Seeing the post did not detectably change that morning's trades. Case studies traced the adoption of positions and decision rules over subsequent days, and communicating groups sold in step more often than isolated ones. Because isolated groups also converged on the same stocks, sometimes completely, assessments of herding need to measure excess convergence relative to a no-peer baseline.