Fokker-Planck and Kolmogorov backward equations for continuous time random walk scaling limits (Q2832839)

The paper under review unifies the solutions of the Fokker-Planck equation and aims at understanding the distributions of scaling limits of continuous-time random walks (CTRWs) in terms of integro-differential equations.NEWLINENEWLINESection 2 uses results of [\textit{M. M. Meerschaert} and \textit{P. Straka}, Ann. Probab. 42, No. 4, 1699--1723 (2014; Zbl 1305.60089)] and exploits the Skorokhod continuity of the path mapping to obtain the CTRW scaling limit. With the transience in the time-component, the authors present Theorem 2.2 to characterise the distribution of \(X_t\) for Lebesgue almost every \(t\in \mathbb R\).NEWLINENEWLINESection 3 gives a projection of the semigroup in the proper Banach space framework. Section 4 gives an example of the Kolmogorov backward equation for CTRW limits.NEWLINENEWLINESection 5 shows that the probability law of the CTRW limit is a unique solution of the Fokker-Planck equation as long as the temporal Lévy measure is time-independent or the corresponding operator is invertible. Theorem 5.2 gives the Fokker-Planck equation for CTRW limits. Special cases of Fokker-Planck operators are also given.NEWLINENEWLINESection 6 is devoted to subdiffusions in a time-dependent potential and traps of spatially varying depth as well as space- and time-dependent Lévy walks, to further illustrate the results.