No. 956 - Agent-based modeling at central banks: recent developments and new challenges
This paper provides an overview of central bank research on agent-based simulation models over the past two decades, distinguishing between applied studies, technical developments and direct use in decision-making processes. In addition, a perspective on possible future applications of the methodology to new types of analysis is proposed.
Agent-based models have specific advantages over other methodologies, such as the ability to easily accommodate heterogeneity, non-linearities and out-of-equilibrium dynamics. In central banks, they have initially been used for the study of payment systems, later for analyses of systemic risks and recently for macroeconomic forecasting. The growing availability of data and computational power is increasing their reliability, making them possible analysis tools even for especially complex tasks, such as assessing the potential impacts of climate change or artificial intelligence on the economy.
Full text
-
28 July 2025
YouTube
X - Banca d'Italia
Linkedin
RSS
E-mail Alert
