discretizing a backward stochastic differential equation倒向随机微分方程离散化.pdfVIP

IJMMS 32:2 (2002) 103–116 PII. S0161171202110234 © Hindawi Publishing Corp. DISCRETIZING A BACKWARD STOCHASTIC DIFFERENTIAL EQUATION YINNAN ZHANG and WEIAN ZHENG Received 24 October 2001 We show a simple method to discretize Pardoux-Peng’s nonlinear backward stochastic differential equation. This discretization scheme also gives a numerical method to solve a class of semi-linear PDEs. 2000 Mathematics Subject Classification: 60H10, 60H35, 65M12. 1. Introduction. Given a probability space (Ω,F,P). Let Wt be a standard Brownian motion with (Ft ) ⊂ F as its natural filtration. Given any positive constant T ∞ and a random variable ξ ∈ FT . A backward stochastic differential equation is the equation [9] Y = ξ − T f Y ,Z ,sds − T Z dW , (1.1) t s s s s t t where (Y ,Z ) are unknown predictable processes. We will assume that f is a Lipschitz t t function with respect to its arguments throughout this paper. Since this equation has its important applications into control theory and mathematical finance, many math- ematicians are not satisfied merely by descriptive existence theorems. They are also interested in constructing the numerical solutions. In order to make real construction, Antonelli [ 1] solved in short time the coupled forward-backward stochastic differen- tial equations, in which it is assumed that ξ = g(V ) where {V } is the solution to a T t t forward stochastic differential equation V t ds − t