Analysis of Learning Types in an Artificial Market
- 932 Downloads
In this paper, we examined the conditions under which evolutionary algorithms (EAs) are appropriate for artificial market models. We constructed three types of agents, which are different in efficiency and accuracy of learning. They were compared using acquired payoff in a minority game, a simplified model of a financial market. As a result, when the dynamics of the financial price was complex to some degree, an EA-like learning type was appropriate for the modeling of financial markets.
KeywordsPrice Change High Payoff Agent Type Payoff Matrix Standard Agent
Unable to display preview. Download preview PDF.
- 3.Izumi, K., Nakamura, S., Ueda, K.: Development of an artificial market model based on a field study. Information Science (in press)Google Scholar
- 4.Arthur, W.B.: Inductive reasoning and bounded rationality (the el farol problem). American Economic Review 84, 406 (1994)Google Scholar
- 5.Minority Game’s web page, http://www.unifr.ch/econophysics/
- 6.Zhang, Y.-C.: Modeling market mechanism with evolutionary games. Europhys. News 29, 51–54 (1998)Google Scholar
- 11.Joshi, S., Parket, J., Bedau, M.A.: Technical trading creates a prisoner’s dilemma: Results from an agent-based model. In: Abu-Mostafa, Y.S., LeBaron, B., Lo, A.W., Weigend, A.S. (eds.) Computational Finance 1999, pp. 465–479. MIT Press, Cambridge (2000)Google Scholar
- 14.Chen, S.H., Liao, C.C.: Agent-based computational modeling of the stock price-volume relation. Information Sciences (in press)Google Scholar