For a corporation in global supply chain, transfer prices, retail prices and order quantities are often the fundamental decision variables to obtain the maximal after-tax profit. Taking into consideration the uncertainty of demand, cost of production, allocation of the transportation cost, shortage loss, tax rates and limitation on markdown rates, this paper presents an optimization model to formulate the problem of the global supply chain management. By expectation method, a deterministic equivalent formulation is first derived for the constructed stochastic model in the case that the demands are continuous random variables. Owing to the complicacy caused by evaluating the integrals with the unknown decision variables in the objective function, an efficient algorithm is developed to solve the model based on the gradient information of the objective function and constraints. By numerical experiments, many managerial implications are revealed from the presented model and algorithm in this paper.