Backward stochastic differential equations with nonlinear Young driver

Abstract: We study backward stochastic differential equations (BSDEs) involving a nonlinear Young integral of the form $\int_{t}^{T}g(Y_{r})\eta(dr,X_{r})$, where the driver $\eta(t,x)$ is a space-time H\"older continuous function and $X$ is a diffusion process. Solutions to such equations provide a probabilistic interpretation of the solutions to stochastic partial differential equations (SPDEs) driven by space-time noise. We first obtain the existence and uniqueness of the solutions to these BSDEs under the assumption that the driver $\eta(t,x)$ is bounded, by a modified Picard iteration method. Then, we derive the comparison principle by analyzing associated linear BSDEs and further establish regularity results for the solutions. Finally, by combining the regularity results with a localization argument, we extend the well-posedness result to the case where $\eta(t,x)$ is unbounded. As applications, we derive nonlinear Feynman-Kac formulae for a class of partial differential equations with Young drivers (Young PDEs), and further investigate BSDEs with non-Lipschitz coefficient functions, as well as linear stochastic heat equations subject to Neumann boundary conditions.