scientific article; zbMATH DE number 4038949
Let, for every \(\epsilon >0\), a Markov process \((x_{\epsilon},y_{\epsilon})\) with infinitesimal generator \(A=A_ 0+\epsilon^{-1} A_ 1\) be given. It is assumed that for every fixed y the operator \(A_ 0\) is the generator of a deterministic motion with velocity given by some function a and for every fixed x the operator \(A_ 1\) is the generator of a certain ergodic jump process. The authors prove that the processes \(x_{\epsilon}\) converge weakly, as \(\epsilon\) tends to zero, to some process x which is a deterministic motion with velocity \(\bar a,\) where \(\bar a\) is an average of the function a with respect to the ergodic measure of the process generated by \(A_ 1.\) In case of \(\bar a\) equal to zero it is shown that the processes \(x_{\epsilon}\) converge weakly to the solution of a certain stochastic differential equation provided the time scale is appropriately changed.
- Dynamical systems under the action of fast random perturbations
- scientific article; zbMATH DE number 1515765
- scientific article; zbMATH DE number 1336702
- Asymptotic methods for stability analysis of Markov dynamical systems with fast variables
- Fluctuation of stochastic systems with average equilibrium point
- An averaging principle for dynamical systems in Hilbert space with Markov random perturbations
- Extended Poisson equation for weakly ergodic Markov processes
- Diffusion approximation of systems with weakly ergodic Markov perturbations. I
- scientific article; zbMATH DE number 90006 (Why is no real title available?)
- Convergence of switching diffusions
- Martingale problems for switched processes
- Quantitative ergodicity for some switched dynamical systems
- Dynamical systems under the action of fast random perturbations
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3777182)