Infinite dimensional BSDE with jumps (Q3146473)

Since the paper of \textit{E. Pardoux} and \textit{S. G. Peng} [Syst. Control Lett. 14, No. 1, 55-61 (1990; Zbl 0692.93064)] on nonlinear backward stochastic differential equations, there has been a lot of authors working on generalizations and applications for this type of equation. Among these generalizations there are extensions to the Hilbert space setting [cf. \textit{E. Pardoux} and \textit{A. Răşcanu}, Stochastics Stochastics Rep. 67, No. 3/4, 159-167 (1999; Zbl 0948.60049)] as well as extensions to backward equations in a Wiener-Poisson context [cf. \textit{G. Barles}, \textit{R. Buckdahn} and \textit{E. Pardoux}, ibid. 60, No. 1/2, 57-83 (1997; Zbl 0878.60036) and \textit{E. Pardoux}, in: Backward stochastic differential equations. Pitman Res. Notes Math. Ser. 364, 207-219 (1997; Zbl 0886.60053)]. The authors of the present paper study backward stochastic differential equations in the setting of reflexive Banach spaces \(V_i\), \(1\leq i\leq q\), which are continuously embedded in a separable Hilbert space \(H\). Their equations are driven by a cylindrical Brownian motion \(W\) defined over another separable Hilbert space \(K\) and a compensated Poisson random measure \(\widetilde\mu\) which intensity measure \(\Pi\) is a \(\sigma\)-finite measure defined over some measurable space \(X\). Under appropriate assumptions on the coefficient \[ G: [0,T]\times \Omega\times \bigcap_{i\leq q} V_i\times L_{H.S.} (K\to H)\times L^2(X\to H;\Pi)\to \Biggl(\bigcap_{i\leq q} V_i\Biggr)' \] and the terminal condition \(\xi\in L^p(\Omega,{\mathcal F}^{W,\mu}_T)\) they prove existence and uniqueness of a solution to the equation \[ dY_t= -G(s, Y_s, Z_s, U_s) ds+ Z_s dW_s+ \int_X U_s(x) \widetilde\mu(ds dx),\qquad Y_T= \xi. \] The proof of existence is based on Galerkin's approximation which uses the existence of an orthonormal basis of \(H\) to reduce the infinite-dimensional equation to finite-dimensional ones. In the last section of their paper the authors give some applications to backward stochastic partial differential equations.