Agent Based Computational Economics: A Review, Challenges And Future Direction

Abstract

Agent-Based Computational Economics (from now ACE) is a dynamic field that combines computational techniques with economic theory to model and analyze complex adaptive systems. It originated from Guy Orcutt's pioneering work in 1957, which introduced microsimulation for economic transactions and interactions. ACE has evolved significantly, particularly with advanced computational technologies in the mid-1990s, leading to the rise of agent-based models (ABMs) and complex adaptive systems (CAS). These advancements allow researchers to simulate individual agents' behaviors and interactions within an economy, revealing emergent properties of economic systems. ACE sets itself apart from classical economic theory by incorporating the concept of bounded rationality, which acknowledges that decision-makers have limited information and cognitive capabilities. This approach also emphasizes the constant interaction among these decision-makers and the existence of multiple equilibrium situations. Overall, it offers a heterodox perspective that diverges from the traditional economic modeling methods. However, ACE studies face some challenges and limits. The main objective of this research is to conduct a literature review, analyze the historical progression of ACE, examine recent advancements and challenges, and explore the potential future trajectory of this economic approach.