TraderTalk: An LLM Behavioural ABM applied to Simulating Human Bilateral Trading Interactions

TL;DR

TraderTalk introduces a hybrid agent-based model using Large Language Models to simulate realistic bilateral trading interactions, capturing nuanced human behaviors in financial markets.

Contribution

This paper presents a novel GABM that integrates LLMs into ABMs to enhance realism in simulating human trading behaviors and interactions.

Findings

Replicates trade-to-order volume ratios in bond markets

Addresses coordination and interpretation challenges in LLM-based agents

Enhances realism through opportunistic prompt design

Abstract

We introduce a novel hybrid approach that augments Agent-Based Models (ABMs) with behaviors generated by Large Language Models (LLMs) to simulate human trading interactions. We call our model TraderTalk. Leveraging LLMs trained on extensive human-authored text, we capture detailed and nuanced representations of bilateral conversations in financial trading. Applying this Generative Agent-Based Model (GABM) to government bond markets, we replicate trading decisions between two stylised virtual humans. Our method addresses both structural challenges, such as coordinating turn-taking between realistic LLM-based agents, and design challenges, including the interpretation of LLM outputs by the agent model. By exploring prompt design opportunistically rather than systematically, we enhance the realism of agent interactions without exhaustive overfitting or model reliance. Our approach successfully replicates trade-to-order volume ratios observed in related asset markets, demonstrating the potential of LLM-augmented ABMs in financial simulations