Two-stage grey support vector machine prediction model

Huaping Zhou , Yue Yuan

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12C) 372-378

Faculty of Computer Science & Engineering Anhui University of Science and Technology, HuaiNan, China

Two-stage grey support vector machine prediction model (D2GM-SVM) is put forward by analysing the grey model GM , support vector machine model (SVM) and one-stage grey support vector machine prediction model((DGM-SVM). The prediction accuracy of grey model is improved through two-stage buffer operators D2 to predict the various relevant indexes. At the same time, genetic algorithm is used to find the optimal parameters of the support vector machine model, RBF kernel parameter and penalty parameter, which are the optimal parameters (c, g). Thus, the regression model of the optimal support vector machine is determined. Finally, the final output value is predicted by inputting the predictive value of each index into the support vector machine model. The results show D2GM-SVM has a higher prediction accuracy compared with grey prediction model , BP neural network prediction model and DGM-SVM in this case, and that grey forecasting model combined with the support vector machine model has practical value in solving practical prediction problems.