Asset Price Dynamics among Heterogeneous Interacting Agents

  • Authors:
  • Carl Chiarella;Mauro Gallegati;Roberto Leombruni;Antonio Palestrini

  • Affiliations:
  • School of Finance and Economics, University of Technology, Sydney, Australia/ E-mail: Carl.Chiarella@uts.edu.au/;Department of Economics, University of Ancona, Italy/ E-mail: gallegati@dea.unian.it;Laboratorio R. Revelli, Torino, Italy/ E-mail: rleom@rocketmail.com;Department of Economics, University of Ancona, Italy/ E-mail: antoniop@dea.unian.it

  • Venue:
  • Computational Economics
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we investigate the presence of rationalherding on asset price dynamics during the intra-day trading withheterogeneous interacting agents, whose information set is notcomplete. In the model, individual probability measures offinancial investment strategies are defined using statisticalmechanics concepts. In addition, there is a learning processtoward the best strategy, implemented as a geneticalgorithm. Simulations show that imitative behavior can be arational strategy, since it allows an investor to gain excessreturns on an asset by exploiting information regarding pricedynamics not strictly contained in the fundamental solution. Herdbehavior is rational in the sense that it produces profits at theexpense of increasing the complexity of the system.