A mechanism for pockets of predictability in complex adaptive systems

¹ UFR de Sciences Economiques, Gestion, Mathématiques et Informatique CNRS UMR 7536 and Université Paris X-Nanterre - 92001 Nanterre Cedex, France
² Laboratoire de Physique de la Matière Condensée, CNRS UMR 6622 and Université de Nice-Sophia Antipolis - 06108 Nice Cedex 2, France
³ Institute of Geophysics and Planetary Physics and Department of Earth and Space Sciences, University of California - Los Angeles, CA 90095, USA

Corresponding authors: vitting@unice.fr sornette@moho.ess.ucla.edu

Received: 3 January 2005
Accepted: 12 April 2005

Abstract

We document a mechanism operating in complex adaptive systems leading to dynamical pockets of predictability (“prediction days”), in which agents collectively take predetermined courses of action, transiently decoupled from past history. We demonstrate and test it out of sample on synthetic minority and majority games as well as on real financial time series. The surprising large frequency of these prediction days implies a collective organization of agents and of their strategies which condense into transitional herding regimes.