Averaging principle for diffusion processes via Dirichlet forms
Probability
2014-03-27 v2
Abstract
We study diffusion processes driven by a Brownian motion with regular drift in a finite dimension setting. The drift has two components on different time scales, a fast conservative component and a slow dissipative component. Using the theory of Dirichlet form and Mosco-convergence we obtain simpler proofs, interpretations and new results of the averaging principle for such processes when we speed up the conservative component. As a result, one obtains an effective process with values in the space of connected level sets of the conserved quantities. The use of Dirichlet forms provides a simple and nice way to characterize this process and its properties.
Cite
@article{arxiv.1307.4248,
title = {Averaging principle for diffusion processes via Dirichlet forms},
author = {Florent Barret and Max-K. Von Renesse},
journal= {arXiv preprint arXiv:1307.4248},
year = {2014}
}
Comments
31 pages