A note on the reflected backward stochastic differential equations driven by a Lévy process with stochastic Lipschitz condition (Q426974): Difference between revisions

Given a proper lower semi-continuous convex function \(\Phi:\mathbb{R}\rightarrow (-\infty,+\infty]\) and the family of orthogonal \(i\)th-power jump martingales \(H^{(i)},\;i\geq 1\), obtained by orthogonalisation from the Teugels martingales associated to a Lévy process, the authors of the present paper study a reflected BSDE of the form \[ dY_t=-f(t,Y_t,Z_t)dt-dK_t+\sum_{i\geq 1}Z_t^{(i)})dH^{(i)}_t,\;Y_T=\xi, \] where, for the solution \((Y,Z,K)\), the process \(Y\) is supposed to take its values in \(\overline{\operatorname{Dom}(\Phi)}\) and \(dK_t=k_tdt\) is an absolutely continuous process such that, for a given process \(a\) defined by the Lipschitz ``constants'', which are in fact stochastic processes here, \(-k_t\in a^2(t)\partial\Phi(Y_t)\). The authors prove the existence and the uniqueness for this BSDE using the Yosida approximation of the subdifferential \(\partial\Phi\) of \(\Phi\). The authors' approach is a direct combination of earlier works by Pardoux and Rascanu (1998) on BSDEs with subdifferential operator (backward stochastic variational inequalities), by Nualart and Schoutens (2001) on BSDEs and the Feynman-Kac formula for Lévy processes and, for instance, by Bender and Kohlmann who studied BSDEs with stochastic Lipschitz condition, i.e., BSDEs which driving coefficient satisfies a Lipschitz condition in which the Lipschitz constants are replaced by adapted stochastic processes.