Ergodic theorems and exponential decay of sample paths for certain interacting diffusion systems (Q2367027): Difference between revisions

The author studies the ergodic behaviour of a class of interacting diffusion processes of configurations in \(\mathbb{Z}^ d\), governed by stochastic differential equations of the form \[ dx_ i(t)=\sum_{j \in \mathbb{Z}^ d}A_{ij}dt+a(x_ i(t))dB_ i(t)\qquad (i \in \mathbb{Z}^ d), \] which appear in various cases (e.g. stepping stone models with random drift or selection, branching diffusion, scalar field in a non-stationary random potential, Ornstein-Uhlenbeck type processes). For \(\mathbb{Z}^ d\)- translation invariant \((A_{ij})\) he describes under suitable assumptions their \(\mathbb{Z}^ d\)-translation invariant stationary measures and proves convergence to them for \(\mathbb{Z}^ d\)-translation invariant initial distributions. He furthermore derives an assumption for the exponential decay of the sample paths. The mentioned examples are covered by the respective assumptions.