Strong predictor-corrector Euler-Maruyama methods for stochastic differential equations with Markovian switching (Q455819)

A family of predictor-corrector Euler-Maruyama numerical methods (PCEM) to approximate the solution of the stochastic differential equation with Markovian switching (SDEwMS) NEWLINE\[NEWLINEdy= f(y(t), r(t))\,dt+ g(y(t), r(t))\,dW(t),\quad t\geq 0,NEWLINE\]NEWLINEis presented. Under global Lipschitz and linear growth conditions on the drift coefficient, the diffusion coefficient, and the corrected drift function, strong convergence with order 0.5 to the exact solution is proved for PCEM. The extension to the multidimensional SDEwMS is described. A criterion is developed to determine numerical \(p\)-stability for a linear test SDEwMS. Data is given that verifies that accurate approximations of the known solution are produced by several versions of the PCEM for an example of an appropriate linear test SDEwMS.