Agent-Based Modelling for Real-World Stock Markets under Behavioral Economic Principles

TL;DR

The paper tackles the challenge of realistically simulating stock-market dynamics and enabling counterfactual analyses by combining agent-based modelling with behavioral-economic trader rules and a neural calibration pipeline. It leverages a variational autoencoder to infer agent parameters from limit order book data, and uses a surrogate model to connect these parameters to stylized facts, addressing non-differentiability in market dynamics. A key contribution is aligning the ABM with public economic indices (e.g., CPI) to boost explainability, supported by empirical evidence that the simulator reproduces stylized facts with high fidelity and delivers efficient inference compared to standard baselines. The work demonstrates meaningful correlations between calibrated agent parameters and macro indices, highlighting potential for interpretable scenario analysis and strategy testing in real-world markets.

Abstract

The reproduction of realistic dynamics in financial markets is of great significance, as it enhances our understanding of market evolution beyond other physical processes, and facilitates the development and backtesting of investment strategies. Most existing literature approaches this issue as a time series forecasting problem, which often faces challenges such as 1) overfitting historical data, 2) failing to reconstruct stylized facts, and 3) limiting users' ability to conduct counterfactual analyses. To address these limitations, we employ agent-based modeling (ABM) for market simulation, where each trader acts as an autonomous agent guided by established behavioral-economic principles. The parameters of the agent model are subsequently calibrated using deep learning techniques. Additionally, we align our agent model with publicly available economic indices, such as the Consumer Price Index (CPI), to enhance the explainability of our system's outcomes. Our experiments demonstrate that the ABM method effectively reproduces market dynamics with a confidence level of 90%, accurately reflecting well-known stylized facts. Furthermore, the calibration process proves to be more computationally efficient compared to other existing methods that perform simulation-based inference. We also present case studies illustrating the correlation between agent parameters and economic indices.

Figures (14)

• Figure 1: An illustration of the entire ecology. This work focuses on the part of agent-based modelling and calibration.
• Figure 2: An illustration of the limit order book.
• Figure 3: The pipeline of calibrating our agent-based models.
• Figure 4: An illustration of how prices and order sizes are correlated under the CARA utility function, with $\hat{P}_t = 10$, $\Delta m_t = 1$, and sampled $\beta$ varying from $1$ to $3$.
• Figure 5: An illustration of the simulated outcome for half of a trading day.