Recent works have increasingly applied Large Language Models (LLMs) as agents in financial stock market simulations to test if micro-level behaviors aggregate into macro-level phenomena. However, a crucial question arises: Do LLM agents’ behaviors align with real market participants? This alignment is key to the validity of simulation results. To explore this, we select a financial stock market scenario to test behavioral consistency. Investors are typically classified as fundamental or technical traders, but most simulations fix strategies at initialization, failing to reflect real-world trading dynamics. In this work, we assess whether agents’ strategy switching aligns with financial theory, providing a framework for this evaluation. We operationalize four behavioral-finance drivers—loss aversion, herding, wealth differentiation, and price misalignment—as personality traits set via prompting and stored long-term. In year-long simulations, agents process daily price-volume data, trade under a designated style, and reassess their strategy every 10 trading days. We introduce four alignment metrics and use Mann–Whitney U tests to compare agents’ style-switching behavior with financial theory. Our results show that recent LLMs’ switching behavior is only partially consistent with behavioral-finance theories, highlighting the need for further refinement in aligning agent behavior with financial theory.
@inproceedings{li-etal-2026-behavioral,
title = "Behavioral Consistency Validation for {LLM} Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation",
author = "Li, Zeping and
Wan, Guancheng and
Chen, Keyang and
Chen, Yu and
Zhao, Yiwen and
Torr, Philip and
Ye, Guangnan and
Yin, Zhenfei and
Chai, Hongfeng",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.2006/",
doi = "10.18653/v1/2026.findings-acl.2006",
pages = "40356--40370",
ISBN = "979-8-89176-395-1",
abstract = "Recent works have increasingly applied Large Language Models (LLMs) as agents in financial stock market simulations to test if micro-level behaviors aggregate into macro-level phenomena. However, a crucial question arises: Do LLM agents' behaviors align with real market participants? This alignment is key to the validity of simulation results. To explore this, we select a financial stock market scenario to test behavioral consistency. Investors are typically classified as fundamental or technical traders, but most simulations fix strategies at initialization, failing to reflect real-world trading dynamics. In this work, we assess whether agents' strategy switching aligns with financial theory, providing a framework for this evaluation. We operationalize four behavioral-finance drivers{---}loss aversion, herding, wealth differentiation, and price misalignment{---}as personality traits set via prompting and stored long-term. In year-long simulations, agents process daily price-volume data, trade under a designated style, and reassess their strategy every 10 trading days. We introduce four alignment metrics and use Mann{--}Whitney U tests to compare agents' style-switching behavior with financial theory. Our results show that recent LLMs' switching behavior is only partially consistent with behavioral-finance theories, highlighting the need for further refinement in aligning agent behavior with financial theory."
}
%0 Conference Proceedings
%T Behavioral Consistency Validation for LLM Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation
%A Li, Zeping
%A Wan, Guancheng
%A Chen, Keyang
%A Chen, Yu
%A Zhao, Yiwen
%A Torr, Philip
%A Ye, Guangnan
%A Yin, Zhenfei
%A Chai, Hongfeng
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F li-etal-2026-behavioral
%X Recent works have increasingly applied Large Language Models (LLMs) as agents in financial stock market simulations to test if micro-level behaviors aggregate into macro-level phenomena. However, a crucial question arises: Do LLM agents’ behaviors align with real market participants? This alignment is key to the validity of simulation results. To explore this, we select a financial stock market scenario to test behavioral consistency. Investors are typically classified as fundamental or technical traders, but most simulations fix strategies at initialization, failing to reflect real-world trading dynamics. In this work, we assess whether agents’ strategy switching aligns with financial theory, providing a framework for this evaluation. We operationalize four behavioral-finance drivers—loss aversion, herding, wealth differentiation, and price misalignment—as personality traits set via prompting and stored long-term. In year-long simulations, agents process daily price-volume data, trade under a designated style, and reassess their strategy every 10 trading days. We introduce four alignment metrics and use Mann–Whitney U tests to compare agents’ style-switching behavior with financial theory. Our results show that recent LLMs’ switching behavior is only partially consistent with behavioral-finance theories, highlighting the need for further refinement in aligning agent behavior with financial theory.
%R 10.18653/v1/2026.findings-acl.2006
%U https://aclanthology.org/2026.findings-acl.2006/
%U https://doi.org/10.18653/v1/2026.findings-acl.2006
%P 40356-40370
[Behavioral Consistency Validation for LLM Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation](https://aclanthology.org/2026.findings-acl.2006/) (Li et al., Findings 2026)