scientific article; zbMATH DE number 124388

From MaRDI portal
Publication:4026012





[For the entire collection see Zbl 0746.00063.] Let \(\lambda\in(0,1]\) be fixed and consider a generic, smooth, symmetric matrix valued mapping \(\alpha:\mathbb{R}^ N\to\mathbb{R}^ N\times\mathbb{R}^ N\) which satisfies \[ \lambda\cdot I_{\mathbb{R}^ N}\leq\alpha(x)\leq\lambda^{-1}\cdot I_{\mathbb{R}^ N},\qquad x\in\mathbb{R}^ N, \] in the sense of nonnegative matrices. It is well-known that each of the elliptic differential operators (in the following denoted by \(L)\) \[ \alpha\nabla^ 2=\sum^ N_{i,j=1}\alpha^{i,j}(x){\partial^ 2\over\partial x^ i\partial x^ j}\qquad\text{and }\quad \nabla(\alpha\nabla)=\sum^ N_{i,j=1}{\partial\over\partial x^ i}\left(\alpha^{i,j}(x){\partial\over\partial x^ j}\right) \] determines a unique Markov semigroup \(P^ L\) on \(\mathbb{C}_ b(\mathbb{R}^ N,\mathbb{R})\) given by \[ \bigl(P^ L_ t\varphi\bigr)(x)=\int_{\mathbb{R}^ N}\varphi(y)P^ L(t,x,dy),\qquad (t,x)\in(0,\infty)\times\mathbb{R}^ N, \] for every \(\varphi\in\mathbb{C}_ b(\mathbb{R}^ N,\mathbb{R})\). The aim of this excellent overview is to report on a few estimates which the transition probability function \(P^ L(t,x,.)\) is known to satisfy. The author describes a way to prove these bounds and gives historical comments. The methodology presented in the paper is applicable to a variety of related situations.











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 Q4026012)