Academia.eduAcademia.edu

Multistability and cyclic attractors in duopoly games

2000, Chaos, Solitons & Fractals

Abstract

A dynamic Cournot duopoly game, whose time evolution is modeled by the iteration of a map X xY y 3 r 1 yY r 2 x, is considered. Results on the existence of cycles and more complex attractors are given, based on the study of the one-dimensional map p x r 1 r 2 x. The property of multistability, i.e. the existence of many coexisting attractors (that may be cycles or cyclic chaotic sets), is proved to be a characteristic property of such games. The problem of the delimitation of the attractors and of their basins is studied. These general results are applied to the study of a particular duopoly game, proposed in M. Kopel [Chaos, Solitons & Fractals, 7 (12) (1996) 2031±2048] as a model of an economic system, in which the reaction functions r 1 and r 2 are logistic maps. Ó

Key takeaways

  • General properties of maps T : xY y® r 1 y Y r 2 x A trajectory of the map T starting from an i.c.
  • In fact, as shown in [1], whenever a bifurcation occurs that creates (eliminates) cycles of the map p, and thus also of q, many cycles of the Cournot map are simultaneously created (eliminated) at the same parameter's value.
  • In general all the attractors of , cycles and cyclic chaotic sets, and their local and global bifurcations, can be obtained from the knowledge of the dynamics of the map p. As an example, in the next section we shall consider the local and global bifurcations of the map p de®ned in (9), from which all the dynamical properties of the Cournot game (5) proposed in [10] can be deduced.
  • For the map (9) a necessary condition for the existence of two coexisting distinct attractors is that p has four ®xed points.
  • Consider for example l 3X8567 at which only 3-cyclic chaotic intervals exist for p. In this case has two coexisting attractors: a 3cyclic chaotic rectangle and a 6-cyclic chaotic rectangle, in the absorbing square t  t X Invariant (and chaotic) one-dimensional sets belonging to these trapping regions are shown in Figs.