FOUNDATIONAL PROBLEMS OF SIMULATION APPROACHES TO ECONOMICS EXEMPLIFIED THROUGH SWARM MODELS

That there are foundational problems involved in simulation approaches to economics can hardly come as a surprise to the reader. Some of these problems are inherited from economics but are exposed anew when a simulation approach is chosen, and computability required. These are problems as defining value and the purpose of economic activity and handling uncertainty. Other problems are related to simulation rather than economics. Simulation models within all sciences bear the risk of becoming "one damned thing after another" as Putnam has phrased it, i.e. there is a problem of disciplining or anchoring a science that makes use of synthesis as its central method.One way of overcoming part of the simulation related foundational problems, is by increasing communication between modelers and making it easier to replicate models and to reuse programming code. Developing and understanding simulation models would be much easier if we had a few well-tried trade algorithms, bankruptcy algorithms etc. to use and refer to. Obtaining such advantages has been one of the goals of the Swarm platform.The purpose of the presented paper is to identify foundational problems of simulation approaches to economics, originating from economics as well as simulation, and go through a group of economic Swarm models to see how these problems are dealt with by each individual model. Next, the purpose is to look for elements that may be standardized for common use. A more superior purpose, or hope, is to contribute to increased communication between Swarm modelers.