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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Vidler, Alicia, Walsh, Toby
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913685745696768
author Vidler, Alicia
Walsh, Toby
author_facet Vidler, Alicia
Walsh, Toby
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2410_21280
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TraderTalk: An LLM Behavioural ABM applied to Simulating Human Bilateral Trading Interactions
Vidler, Alicia
Walsh, Toby
Trading and Market Microstructure
Artificial Intelligence
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
title TraderTalk: An LLM Behavioural ABM applied to Simulating Human Bilateral Trading Interactions
topic Trading and Market Microstructure
Artificial Intelligence
url https://arxiv.org/abs/2410.21280