AI agents in trading systems exhibit phase transitions and altered market impact

A new research paper explores the behavior of Limit Order Books (LOBs) when populated solely by autonomous reinforcement learning agents. The study identifies distinct phase transitions in order flow, separating stable price discovery from volatile cascade states based on agent numbers and market depth. It also reveals that market impact dynamics under agentic liquidity provision differ from traditional models, showing unique regimes influenced by non-linear feedback loops. AI

IMPACT This research could inform the development of more stable and predictable algorithmic trading systems by understanding agent-driven market dynamics.

RANK_REASON The cluster contains a research paper detailing novel findings in quantitative finance using AI agents. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.AI →

• Agentic Limit Order Books
• market impact
• order-matching engine
• Price discovery
• reinforcement learning
• Traders

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →