Harnack inequality for SDE with multiplicative noise and extension to Neumann semigroup on nonconvex manifolds (Q717885): Difference between revisions

Consider the following stochastic differential equation in \(\mathbb{R}^{d}\) \[ dX_{t}=\sigma\left( t, X_{t} \right) dB_{t}+b\left( t, X_{t} \right) dt, \tag{1} \] where \(B_{t}\) is a \(d\)-dimensional Brownian motion on a complete probability space. Consider the operator \(P_{t}\) on the space of bounded non-negative measurable functions on \(\mathbb{R}^{d}\) defined by \[ P_{t}f\left( x \right):=Ef\left( X_{t}^{x} \right), t \geqslant 0, x \in \mathbb{R}^{d}, \] where \(X_{t}^{x}\) is the solution to (1) with \(X_{0}=x\). The main results of the paper include what is known as log-Harnack and Wang's Harnack inequalities for \(P_{t}\) under certain conditions on \(b\) and \(\sigma\). The applications of these inequalities described in the paper are upper heat kernel estimates and extensions of these inequalities to a reflecting diffusion on a Riemannian manifold with non-convex boundary.