Convergence rate of numerical solutions to SFDEs with jumps (Q645694)

The authors consider stochastic functional differential equations (SFDEs) with jumps of the form \[ dx(t)=f(x_{t})dt+g(x_{t})dB_{t}+h(x_{t})dN(t),\;0\leq t\leq T, \] with given \(x_{0}\),\ where \(x\) is \(n\)-dimensional,\(\;x_{t}:=\{x(t+\theta ),\;-\tau\leq\theta\leq0\}\), \(x_{t^{-}}:=\{x((t+\theta)^{-}),\;-\tau\leq \theta\leq0\}\),\(\;\;\;x(t^{-}):=\lim_{s\uparrow t}x(s)\), \(B(t)\) is an \(m\)-dimensional Brownian motion, and \(N(t)\) is a scalar Poisson process. Under a global Lipschitz condition they show that the \(p\)th-moment convergence of Euler-Maruyama numerical solutions to SFDEs with jumps has the order \(1/p\) for any \(p\geq2\). This is different from the case of SFDEs without jumps, where the order is \(1/2\) for any \(p\geq2\). They consider also the mean-square convergence under a local Lipschitz condition.