AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

MarketSim: Simulating Stock Markets with Large-Scale Generative Agents

Jinghua Piao, zhentao liu, Cheng Huang, Jiarui Huang, Songwei Li, Ranran Wang, Yong Li

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Agent data synthesis and behavior studies

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes MarketSim, a stock-market simulator built on large-scale LLM-based generative agents, aiming to jointly capture structural and behavioral fidelity of market dynamics.

Problem

Existing stock market simulators face a trade-off between structural fidelity (realistic market mechanics and price formation) and behavioral fidelity (realistic participant behavior among heterogeneous agents).

Approach

Uses large-scale generative (LLM-based) agents to simulate heterogeneous market participants interacting in an evolving information landscape, so that prices emerge from decentralized interactions rather than being imposed.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use empirically grounded agent profiles instead of simplistic archetypes (ablation)MarketSim · JNJ tariff shockMAPE: 0.82% to over 2.8%Improved
add self-reflection to manager agents (ablation)MarketSim · stock simulation ablationRMSE: increases RMSE by over 10xImproved
add self-reflection to manager agents (ablation)MarketSim · stylized factsstylized fact reproductionImproved
use a stronger LLM for the manager (ablation)MarketSim · stock simulation ablationprice fidelityImproved
allow dynamic strategy allocation by the manager (ablation)MarketSim · stock simulation ablationprice fidelityImproved
use MarketSim instead of strongest alternative baselinesMarketSim · all stocks and scenariosRMSE: 11%-41%Improved
use MarketSim instead of strongest alternative baselinesMarketSim · all stocks and scenariosMAPE: 24%-39%Improved
use MarketSim instead of strongest alternative baselinesMarketSim · all stocks and scenariosvolatility similarity: 15%-43%Improved
use ABM Transformer instead of MarketSimABM baseline · JNJ tariff shockDTW distance: 0.011 -> 0.005Hurt
integrate predictive models into ABIDES instead of using them standaloneLinear · JNJ tariff shockQ-Q correlation: 0.892 -> 0.988Improved
compare agent decisions against matched control decisionsMarketSim agents · human evaluationmarket consistency t-test: t = 7.09, p < 0.001Improved
add 200 momentum-based agentsMarketSim · disruption experimentmarket volatilityMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy