Stochastic maximum principle for sub-diffusions and its applications

Abstract: In this paper, we study optimal stochastic control problems for stochastic systems driven by non-Markov sub-diffusion $B_{L_t}$, which have the mixed features of deterministic and stochastic controls. Here $B_t$ is the standard Brownian motion on $R$, and $L_t:= \inf{r>0: S_r>t}, \quad t\geq 0,$ is the inverse of a subordinator $S_t$ with drift $\kappa >0$ that is independent of $B_t$. We obtain stochastic maximum principles (SMP) for these systems using both convex and spiking variational methods, depending on whether the convex domain is convex or not. To derive SMP, we first establish a martingale representation theorem for sub-diffusions $B_{L_t}$, and then use it to derive the existence and uniqueness result for the solutions of backward stochastic differential equations (BSDEs) driven by sub-diffusions, which may be of independent interest. We also derive sufficient SMPs. Application to a linear quadratic system is given to illustrate the main results of this paper.