Dynamical systems with fast switching and slow diffusion: Hyperbolic equilibria and stable limit cycles

TL;DR

This paper investigates the long-term behavior of stochastic differential equations with fast Markovian switching and small diffusion, showing convergence of invariant measures to those of an averaged system with limit cycles.

Contribution

It extends the analysis of stochastic systems with fast switching and small noise, proving convergence of invariant measures to the averaged system's limit cycle under hyperbolic equilibrium conditions.

Findings

Invariant measures converge to the averaged system's measure as parameters go to zero.

Results apply to state-dependent switching with bounded, Lipschitz generator.

Includes application to a predator-prey ecological model.

Abstract

We study the long-term qualitative behavior of randomly perturbed dynamical systems. More specifically, we look at limit cycles of stochastic differential equations (SDE) with Markovian switching, in which the process switches at random times among different systems of SDEs, when the switching is fast and the diffusion (white noise) term is small. The system is modeled by $$ dX^{\epsilon,\delta}(t)=f(X^{\epsilon,\delta}(t), \alpha^\epsilon(t))dt+\sqrt{\delta}\sigma(X^{\epsilon,\delta}(t), \alpha^\epsilon(t))dW(t) , \ X^\epsilon(0)=x, $$ where $\alpha^\epsilon(t)$ is a finite state space Markov chain with irreducible generator $Q=(q_{ij})$. The relative changing rates of the switching and the diffusion are highlighted by the two small parameters $\epsilon$ and $\delta$. We associate to the system the averaged ODE \[ d\bar X(t)=\bar f(\bar X(t))dt, \ X(0)=x, \] where $\bar f(\cdot)=\sum_{i=1}^{m_0}f(\cdot, i)\nu_i$ and $(\nu_1,\dots,\nu_{m_0})$ is the unique invariant probability measure of the Markov chain with generator $Q$. Suppose that for each pair $(\epsilon,\delta)$ of parameters, the process has an invariant probability measure $\mu^{\epsilon,\delta}$, and that the averaged ODE has a limit cycle in which there is an averaged occupation measure $\mu^0$ for the averaged equation. We are able to prove that if $\bar f$ has finitely many unstable or hyperbolic fixed points, then $\mu^{\epsilon,\delta}$ converges weakly to $\mu^0$ as $\epsilon\to 0$ and $\delta \to 0$. Our results generalize to the setting of state-dependent switching \[ \mathbb{P}\{\alpha^\epsilon(t+\Delta)=j~|~\alpha^\epsilon=i, X^{\epsilon,\delta}(s),\alpha^\epsilon(s), s\leq t\}=q_{ij}(X^{\epsilon,\delta}(t))\Delta+o(\Delta),~~ i\neq j \] as long as the generator $Q(\cdot)=(q_{ij}(\cdot))$ is bounded, Lipschitz, and irreducible for all $x\in\mathbb{R}^d$. We conclude our analysis by studying a predator-prey model.