An Agent-based Microeconomic Simulation of a local Market

We describe an agent-based microeconomic simulation of a local market. Consumers and Sellers are represented as agents in the Aglets framework. Consumers follow a Cobb-Douglas utility function, and they choose the seller that best matches their desired combination of goods and services. Different sellers provide different combinations of goods and services. The amount of money available to consumers for spending depends on an external parameter representing the state of the economy. Sellers make profit if revenues exceed costs. Although important aspects of a real economy, such as the total available money supply, the equilibrium between supply and demand, and the effect of the labour market, are not explicitly represented in this simple model, they can be incorporated in the future. The advantages of an agent-based simulation are the following: global macroeconomic behaviors arise naturally as a result of local agent interactions; individual consumer and seller agents can become as sophisticated as desired (for example, spatial locations of consumers and sellers may influence consumer choice of a seller, or consumers can follow an income and preference distribution that reflects a particular city), with no constraints imposed by tractability of any mathematical models; the mobile agents (Aglets) framework allows a distributed implementation of the economy over a network of workstations, thus allowing experiments on standard computing hardware with no requirement for access to a supercomputer. Future research will expand on these advantages, to demonstrate the usefulness of the proposed framework for business planning (as a tool to answer what-if questions to a prospective seller), as a teaching tool in economics and business, and as a framework for electronic commerce, whereby real stores and real consumers are represented by agents, who carry out business transactions on their owners’ behalf.