From Heterogeneous expectations to exchange rate dynamic

Philippe Protin, Luc Neuberg and Christine Louargant

No 310, Computing in Economics and Finance 2004 from Society for Computational Economics

Abstract: The purpose of this paper is to analyze how heterogeneous behaviors of agents influence the exchange rates dynamic in the short and long terms. We examine how agents use the information and which kind of information, in order to take theirs decisions to form an expectation of the exchange rate. We investigate a methodology based on interactive agents simulations, following the Santa Fe Artificial Stock Market. Each trader is modeled as an autonomous, interactive agent and the aggregation of their behavior results in foreign exchange market dynamic. Genetic algorithm is the tool used to compute agents, and the simulated market tends to replicate the real EUR/USD exchange rate market. We consider six kinds of agents with pure behavior: fundamentalists, positive feedback traders and negative ones, naive traders, news traders (positive and negative). To reproduce stylized facts of the exchange rates dynamic, we conclude that the key factor is the correct proportion of each agents type, whiteout any need of mimetic behaviors, adaptive agents or pure noisy agents

Keywords: exchange rates dynamic; heterogeneous interactive agents behaviour; genetic algorithm; learning process (search for similar items in EconPapers)
JEL-codes: C63 D83 F31 D84 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-ifn
Date: 2004-08-11

Downloads: (external link)
http://repec.org/sce2004/up.21766.1078078186.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:sce:scecf4:310

Access Statistics for this paper

More papers in Computing in Economics and Finance 2004 from Society for Computational Economics
Contact information at EDIRC.
Series data maintained by Christopher F. Baum ().