Building and assurance of agent-based models: an example and challenge to the field
Author: Midgley, David ; Marks, Robert ; Kunchamwar, DineshINSEAD Area: MarketingIn: Journal of Business Research, vol. 60, no. 8, August 2007 Language: EnglishDescription: p. 884-893.Type of document: INSEAD ArticleNote: Please ask us for this itemAbstract: The assurance, that is, the verification and validation, of agent-based models is difficult, because of the heterogeneity of agents, and the possibility of the emergence of new patterns of macro behavior as a result of the interactions of these agents at the micro-level. This paper uses an agent-based model of the complex interactions among consumers, retailers, and manufacturers to explore issues of model assurance. These explorations involve two challenges for the agent-based model's field. The first challenge is to address the critical issue of software verification. The second challenge is to overcome the many methodological obstacles that exist in empirically validating these models. This paper will outline some of them. The authors propose a method based on the Genetic Algorithm to address both these challenges, but the experiments required, and a lack of good data on many kinds of agents, generally call for a minimalist approach to building and assuring agent-based models.Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
![]() |
Europe Campus | Available | BC007911 |
Ask Qualtrics
The assurance, that is, the verification and validation, of agent-based models is difficult, because of the heterogeneity of agents, and the possibility of the emergence of new patterns of macro behavior as a result of the interactions of these agents at the micro-level. This paper uses an agent-based model of the complex interactions among consumers, retailers, and manufacturers to explore issues of model assurance. These explorations involve two challenges for the agent-based model's field. The first challenge is to address the critical issue of software verification. The second challenge is to overcome the many methodological obstacles that exist in empirically validating these models. This paper will outline some of them. The authors propose a method based on the Genetic Algorithm to address both these challenges, but the experiments required, and a lack of good data on many kinds of agents, generally call for a minimalist approach to building and assuring agent-based models.
Digitized
There are no comments for this item.