TwinMarket：用於金融市場的可擴展行為和社會模擬

TwinMarket: A Scalable Behavioral and Social Simulation for Financial Markets

February 3, 2025

作者: Yuzhe Yang, Yifei Zhang, Minghao Wu, Kaidi Zhang, Yunmiao Zhang, Honghai Yu, Yan Hu, Benyou Wang

cs.AI

摘要

社會出現的研究長期以來一直是社會科學的核心關注焦點。傳統建模方法，如基於規則的基於代理的模型（ABMs），難以捕捉人類行為的多樣性和複雜性，尤其是行為經濟學強調的非理性因素。最近，大型語言模型（LLM）代理已經成為社會科學和角色扮演應用中建模人類行為的仿真工具。研究表明，LLMs可以解釋認知偏見、情緒波動和其他非理性影響，從而實現更現實的社會經濟動態模擬。在這項工作中，我們介紹了TwinMarket，一個利用LLMs來模擬社會經濟系統的新型多代理框架。具體而言，我們研究個體行為如何通過互動和反饋機制產生集體動態和新興現象。通過在模擬股市環境中進行實驗，我們展示了個人行為如何觸發群體行為，導致新興結果，如金融泡沫和經濟衰退。我們的方法提供了對個人決策與集體社會經濟模式之間復雜相互作用的寶貴見解。

English

The study of social emergence has long been a central focus in social science. Traditional modeling approaches, such as rule-based Agent-Based Models (ABMs), struggle to capture the diversity and complexity of human behavior, particularly the irrational factors emphasized in behavioral economics. Recently, large language model (LLM) agents have gained traction as simulation tools for modeling human behavior in social science and role-playing applications. Studies suggest that LLMs can account for cognitive biases, emotional fluctuations, and other non-rational influences, enabling more realistic simulations of socio-economic dynamics. In this work, we introduce TwinMarket, a novel multi-agent framework that leverages LLMs to simulate socio-economic systems. Specifically, we examine how individual behaviors, through interactions and feedback mechanisms, give rise to collective dynamics and emergent phenomena. Through experiments in a simulated stock market environment, we demonstrate how individual actions can trigger group behaviors, leading to emergent outcomes such as financial bubbles and recessions. Our approach provides valuable insights into the complex interplay between individual decision-making and collective socio-economic patterns.