A new way of Support Vector Regression (SVR)-a new regression technique based on the structural risk minimization principle is proposed, it has the similar mathematic form as that of support vector classification, and a simplified Sequential Minimal Optimization (SMO) algorithm based on it is given. Compared with the existing algorithms, the simplified one is simpler in mathematic form and much easier to be implemented without loss in space and time efficiency, which is of benefit to the extensive application of SVR.