Local time of additive Lévy process

From MaRDI portal





An additive Lévy process \(\{X_t\), \(t\in {\mathbb R}^n_+\}\) on \({\mathbb R}^d\) is a simple example of a multi-parameter Lévy process, which can be decomposed into \(X_t= \sum_{j =1}^n X^j_{t_j}\), where the \(X_j\)'s are independent one-parameter Lévy processes on \({\mathbb R}^d\) and \(t_j\) is the \(j\)th coordinate of \(t\). The authors give a criterion for the existence of a jointly continuous local time for such processes, viz. of a continuous density function for the multi-parameter occupation measure. In the main theorem, a sufficient condition is given by \(n\times\alpha > d\), where \(\alpha\) is the minimum of the Blumenthal and Getoor's indices of the \(X^j\)'s. The proof is based on a criterion of \textit{S. M. Berman} [Ann. Math. Stat. 41, 1260-1272 (1970; Zbl 0204.50501)], and on Fourier techniques very similar to the one-parameter case in the easy situation \(\alpha > d\). The authors seem not to be aware of the definitive and much more difficult results of \textit{M. T. Barlow} [Ann. Probab. 16, No. 4, 1389-1427 (1988; Zbl 0666.60072)] about this question for one-parameter Lévy processes.











This page was built for publication: Local time of additive Lévy process

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4502360)