Stability analysis of explicit and semi-implicit derivative-free methods for stochastic differential equations

This paper is devoted to investigate the mean-square stability of explicit and semi-implicit derivative-free methods to a class of stochastic differential equations (SDEs). The mean-square stability functions and regions of explicit and semi-implicit numerical approximation schemes are obtained for a linear stochastic differential equation with multiplicative noise. It is proved that the semi-implicit derivative-free method is mean-square stable compare than the explicit counterpart schemes. A numerical experiment is provided to illustrate the theory.