Locally Feller processes and martingale local problems (Q1994912): Difference between revisions

This paper is devoted to the study of a certain type of martingale problems associated to general operators corresponding to processes, which have finite lifetime. Heuristically, a Lévy-type process \(X\) with symbol \(q:\mathbb R^d \times \mathbb R^d \to C\) is a Markov process, which behaves locally like a Lévy process with characteristic exponent \(q(a, \cdot )\), in a neighbourhood of each point \(a \in \mathbb R^d\) . The pseudo-differential operator \(L\) associated to a Lévy-type process is given by, for \(f \in C_c^\infty (\mathbb R^d)\),\(Lf(a): = - \int_{\mathbb R^d} {{e^{ia \cdot \alpha }}q(a,\alpha )\hat f(\alpha )d\alpha } \), where\(\hat f(\alpha ): = {(2\pi )^{ - d}}\int_{\mathbb R^d} {{e^{ - \alpha \cdot a}}f(a)da} \). The authors investigate the following questions, which naturally appear in studying the Lévy-type processes. 1. Does a sequence \({X^{(n)}}\) of Lévy-type processes, having symbols \({q_n}\), converges towards some process, when \({q_n}\) converges to a symbol \(q\)? 2. What can we say about the sequence \({X^{(n)}}\) when the corresponding sequence of pseudo-differential operators \({L_n}\) converges to an operator \(L\)? 3. What could be the appropriate setting when one wants to approximate a Lévy-type process by a family of discrete Markov chains? In the present paper, the authors describe a general method, which serves as the main tool to tackle the difficulties, which appear in applying earlier obtained results. Moreover, it relaxes some of technical restrictions. For this purpose, they analyze sequences of martingale problems associated to large class of operators acting on continuous functions.