Stochastic bifurcation (Q5899840): Difference between revisions

This paper studies the asymptotic behavior of certain nonlinear dynamical systems under the influence of small stochastic perturbations in the form of additive and/or multiplicative white noise. For systems with different time scales the methods of stochastic averaging and of stochastic normal forms are used to reduce the system's equations to lower dimensional ones. For the resulting Markov processes the qualitative behavior is investigated using the densities of the stationary probability measures associated with the corresponding Markov semigroup. In particular, stability properties of the system's rest point are discussed using a.s. Lyapunov exponents and moment properties. As an application, several examples of bifurcations in stochastic systems are analyzed using the methods of extrema of the invariant density, of Lyapunov exponents, and of bifurcations of moment equations. The examples include codimension 1 and 2 bifurcations, and the shift of a bifurcation point due to the presence of noise is observed. The paper provides a brief, but clear overview of the deterministic theory and of the stochastic theory of bifurcation in terms of Markov process theory. It does not consider the bifurcation aspects of stochastic flows and the related stochastic center manifold theory [see e.g. \textit{L. Arnold} and \textit{P. Boxler} in: ``Diffusion processes and related topics in analysis'', M. Pinsky, ed., Birkhäuser (1990)].