BeforeIT.jl: High-Performance Agent-Based Macroeconomics Made Easy (2502.13267v1)

Abstract: BeforeIT is an open-source software for building and simulating state-of-the-art macroeconomic agent-based models (macro ABMs) based on the recently introduced macro ABM developed in [1] and here referred to as the base model. Written in Julia, it combines extraordinary computational efficiency with user-friendliness and extensibility. We present the main structure of the software, demonstrate its ease of use with illustrative examples, and benchmark its performance. Our benchmarks show that the base model built with BeforeIT is orders of magnitude faster than a Matlab version, and significantly faster than Matlab-generated C code. BeforeIT is designed to facilitate reproducibility, extensibility, and experimentation. As the first open-source, industry-grade software to build macro ABMs of the type of the base model, BeforeIT can significantly foster collaboration and innovation in the field of agent-based macroeconomic modelling. The package, along with its documentation, is freely available at https://github.com/bancaditalia/BeforeIT.jl under the AGPL-3.0.

• The paper presents a Julia package that significantly speeds up agent-based macroeconomic simulations, outperforming Matlab implementations.
• It employs optimized algorithms and a modular architecture to support large-scale Monte Carlo simulations and straightforward model extensions.
• The tool enhances reproducibility and collaboration with comprehensive documentation, unit tests, and open-source availability for policy analysis.

High-Performance Agent-Based Macroeconomic Modelling with BeforeIT.jl

This paper presents "BeforeIT.jl," a software package developed to facilitate the creation and simulation of sophisticated macroeconomic agent-based models (ABMs). In particular, the package is inspired by a state-of-the-art macro ABM recently introduced by Poledna et al., which has demonstrated competitive forecasting capabilities alongside its ability to simulate complex economic dynamics such as asset bubbles and financial crises.

BeforeIT.jl is implemented in Julia, which is renowned for its combination of computational efficiency and ease of use. This decision allows the software to deliver a notable performance improvement over previous implementations in Matlab, with benchmark tests showing that BeforeIT.jl executes several orders of magnitude faster. This efficiency is largely attributed to Julia's just-in-time compilation and optimized algorithms, such as dynamic weighted sampling. These improvements render BeforeIT.jl highly suitable for large-scale simulations and applications requiring significant computational power.

The package emphasizes user accessibility and extensibility. It provides a modular architecture that simplifies modifications and extensions of existing models. By utilizing Julia's macro capabilities, the developers ensure that new agent types or model features can be integrated without extensive duplication of code. Additionally, the design supports running extensive Monte Carlo simulations, which are critical for robust economic analysis and forecasting.

BeforeIT.jl further distinguishes itself with a rich set of features designed to enhance reproducibility and collaboration among researchers. It provides comprehensive documentation and an extensive suite of unit tests, ensuring the reliability and accuracy of simulations. Moreover, the tool is open-source and freely available, which promotes broader adoption and innovation in the field of agent-based macroeconomic modelling.

The research implications of BeforeIT.jl are significant. By lowering computational and technical barriers, this tool enables economists to explore complex economic phenomena and test theoretical models more readily. Its capacity for high-resolution simulations of economic sectors and detailed agent interactions sets a foundation for more sophisticated analyses of policy interventions. As empirical data continues to drive economic research, the integration of agent-based models like those supported by BeforeIT.jl will become increasingly critical for understanding economic dynamics and informing policy decisions.

Future developments hold promising potential for BeforeIT.jl. The paper suggests extending calibration scripts to facilitate broader applicability, including the automatic differentiation of model equations, which could optimize sensitivity analyses and parameter estimation. Additionally, integrating models like CANVAS, developed by the Bank of Canada, into BeforeIT.jl could further establish the package as a versatile tool for global macroeconomic research.

In conclusion, BeforeIT.jl provides a robust platform for the development and simulation of macro ABMs, linking advanced computational performance with accessibility and extensibility. By doing so, it invites a collaborative spirit in the exploration of economic dynamics and policy within the research community. The package represents a meaningful step in the evolution of agent-based macroeconomic modelling tools, with the potential to drive significant advancements in both theoretical understanding and practical application.