Paper: arXiv 2412.20138

Abstract

Significant progress has been made in automated problem-solving using societies of agents powered by large language models (LLMs). In finance, efforts have largely focused on single-agent systems handling specific tasks or multi-agent frameworks independently gathering data. However, the multi-agent systems’ potential to replicate real-world trading firms’ collaborative dynamics remains underexplored. TradingAgents proposes a novel stock trading framework inspired by trading firms, featuring LLM-powered agents in specialized roles such as fundamental analysts, sentiment analysts, technical analysts, and traders with varied risk profiles. The framework includes Bull and Bear researcher agents assessing market conditions, a risk management team monitoring exposure, and traders synthesizing insights from debates and historical data to make informed decisions. By simulating a dynamic, collaborative trading environment, this framework aims to improve trading performance. Detailed architecture and extensive experiments reveal its superiority over baseline models, with notable improvements in cumulative returns, Sharpe ratio, and maximum drawdown, highlighting the potential of multi-agent LLM frameworks in financial trading. TradingAgents is available at https://github.com/TauricResearch/TradingAgents.

Complexity vs Empirical Score

  • Math Complexity: 2.5/10
  • Empirical Rigor: 6.0/10 (capped from 8.0: transaction costs not modelled; see penalty flags)
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper focuses on LLM agent architecture and backtesting metrics like Sharpe ratio and cumulative returns, with minimal advanced mathematics beyond basic statistics, making it highly empirical and implementation-focused.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Explore multi-agent LLM framework<br>to replicate trading firm dynamics"] --> B["Methodology<br>Multi-agent system with specialized<br>roles & collaborative debates"]
  B --> C["Data/Inputs<br>Financial market data &<br>historical trading information"]
  C --> D["Computational Process<br>Agent debates & synthesis of<br>fundamental, technical & sentiment insights"]
  D --> E["Key Findings<br>Superior performance vs baselines<br>↑ Cumulative returns, ↑ Sharpe ratio,<br>↓ Maximum drawdown"]
  E --> F["Outcome<br>Validates multi-agent LLM<br>potential in financial trading"]

Cited References (45)

  1. FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making (2024) — link
  2. TradingGPT: Multi-agent system with layered memory and distinct characters for enhanced financial trading performance — → 2309.03736
  3. QuantAgent: Seeking Holy Grail in Trading by Self-Improving Large Language Model (2024) — link
  4. Large Language Model Agent in Financial Trading: A Survey — → 2408.06361
  5. ReAct: Synergizing Reasoning and Acting in Language Models (2023) — link
  6. Can Large Language Models Beat Wall Street? Unveiling the Potential of AI in Stock Selection (2024) — link
  7. FinRobot: An Open-Source AI Agent Platform for Financial Applications using Large Language Models (2024) — link
  8. Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions (2023) — link
and 37 more
  1. MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework (2024) — link
  2. Improving Factuality and Reasoning in Language Models through Multiagent Debate (2023) — link
  3. Evaluation of OpenAI o1: Opportunities and Challenges of AGI (2024) — link
  4. On The Planning Abilities of OpenAI’s o1 Models: Feasibility, Optimality, and Generalizability (2024) — link
  5. Learning to Reason with LLMs - OpenAI O1 Model (2024)
  6. GPT-4 Technical Report (2024) — link
  7. Generative Agents: Interactive Simulacra of Human Behavior (2023) — link
  8. ChatDev: Communicative Agents for Software Development (2024) — link
  9. Teaching Large Language Models to Reason with Reinforcement Learning (2024) — link
  10. Multi-Agent Collaboration: Harnessing the Power of Intelligent LLM Agents (2023) — link
  11. MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning (2024) — link
  12. PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance (2023) — link
  13. FinGPT: Open-Source Financial Large Language Models (2023) — link
  14. Instruct-FinGPT: Financial Sentiment Analysis by Instruction Tuning of General-Purpose Large Language Models (2023) — link
  15. OPT: Open Pre-trained Transformer Language Models (2022) — link
  16. BloombergGPT: A Large Language Model for Finance (2023) — link
  17. XuanYuan 2.0: A Large Chinese Financial Chat Model with Hundreds of Billions Parameters (2023) — link
  18. BBT-Fin: Comprehensive Construction of Chinese Financial Domain Pre-trained Language Model, Corpus and Benchmark (2023) — link
  19. PaLM: Scaling Language Modeling with Pathways (2022) — link
  20. Llama 2: Open Foundation and Fine-Tuned Chat Models (2023) — link
  21. Qwen Technical Report (2023) — link
  22. Baichuan 2: Open Large-scale Language Models (2023) — link
  23. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models (2023) — link
  24. Unveiling the Potential of Sentiment: Can Large Language Models Predict Chinese Stock Price Movements? (2024) — link
  25. Sentiment trading with large language models — → 2412.19245
  26. Can Large Language Models Beat Wall Street? Unveiling the Potential of AI in Stock Selection (2024) — link
  27. LLMFactor: Extracting Profitable Factors through Prompts for Explainable Stock Movement Prediction (2024) — link
  28. FinMem: A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design (2023) — link
  29. A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist (2024) — link
  30. Memory: enduring traces of perceptual and reflective attention (2011) — link
  31. Towards Mitigating Hallucination in Large Language Models via Self-Reflection (2023) — link
  32. Designing Heterogeneous LLM Agents for Financial Sentiment Analysis (2024) — link
  33. TradingGPT: Multi-Agent System with Layered Memory and Distinct Characters for Enhanced Financial Trading Performance (2023) — link
  34. Alpha-gpt: Human-ai interactive alpha mining for quantitative investment — → 2308.00016
  35. Learning to Generate Explainable Stock Predictions using Self-Reflective Large Language Models — → 2402.03659
  36. Integrating Stock Features and Global Information via Large Language Models for Enhanced Stock Return Prediction (2023) — link
  37. Proximal Policy Optimization Algorithms (2017) — link

5 of 45 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.