Agent‐based computational models and generative social science

Abstract

This article argues that the agent-based computational model permits a distinctive approach to social science for which the term "generative" is suitable. In defend ing this terminology, features distinguishing the approach from both "inductive" and "deductive" science are given. Then, the following specific contributions to social science are discussed: The agent-based computational model is a new tool for empirical research. It offers a natural environment for the study of connection ist phenomena in social science. Agent-based modeling provides a powerful way to address certain enduring-and especially interdisciplinary-questions. It allows one to subject certain core theories-such as neoclassical microeconomics-to important types of stress (e.g., the effect of evolving preferences). It permits one to study how rules of individual behavior give rise-or "map up"-to macroscopic regularities and organizations. In turn, one can employ laboratory behavioral research findings to select among competing agent-based ("bottom up") models. The agent-based approach may well have the important effect of decoupling individual rationality from macroscopic equilibrium and of separating decision science from social science more generally. Agent-based modeling offers powerful new forms of hybrid theoretical-computational work; these are particularly relevant to the study of non-equilibrium systems. The agent-based approach invites the interpretation of society as a distributed computational device, and in turn the interpretation of social dynamics as a type of computation. This interpretation raises important foundational issues in social science-some related to intractability, and some to undecidability proper. Finally, since "emergence" *The author is a senior fellow in Economic Studies at The Brookings Institution and a member of the External Faculty of the Santa Fe Institute.

A generic testing framework for agent-based simulation models

“…Unlike MAS, the developers of ABMS are not just computer scientists and software engineers. There is a wide variety of application domains of ABMS from neuroscience (see, eg, Gu¨rcan et al, 2012) to ecology (see, eg, Grimm et al, 2005), from social sciences (see, eg, Epstein, 2007) to economy (see, eg, Windrum et al, 2007), etc and each domain's experts are trying to build their simulation models by themselves. In this sense, the terminology used in testing of ABMS needs to be more understandable and familiar for the experts of these domains.…”

The challenges and scope of theoretical biology

“…MC simulation seek to solve problems where there is a lack of statistical power by generating large data sets compatible with model assumptions. ABM seek to increase the degrees of freedom of individual agents and in this way, represent more realistic models of behavior where agent rules are assumed to percolate into collective dynamics (Epstein, 1999). This approach often presupposes that coarse-graining a system is not desirable, as the process of abstraction can eliminate essential details required to fully explain, or reproduce, the system dynamics.…”

Systems Thinking in the Forest Service: a Framework to Guide Practical Application for Social-Ecological Management in the Enterprise Program

“…Models can range from small, elegant learning tools to large decision-support tools (Bonabeau 2002, Macal andNorth 2007). Phenomena that ABM has been applied to are immense in their breadth: wealth distribution, trade networks, supply chains, unemployment patterns, organizational behaviors, epidemics, traffic congestion, stock market behavior, voting behavior, coalition formation, migration, urban and regional development, social cooperation, the growth and decline of ancient civilizations, and cognitive science (Epstein 1999, Helbing and Balietti 2015, Macal and North 2007.…”

“…Despite grand claims to being a revolutionary new way of doing science (Bonabeau 2002, Helbing and Balietti 2015, Macal and North 2007, Macy and Willer 2002, ABM does have drawbacks. Most importantly is the challenge of encapsulating the full range of human behaviors and responses and then quantifying them into computer code (Bonabeau 2002, Epstein 1999. Different people respond differently in different situationshow do we capture that variation?…”