Abstract: Support vector machine improves the generalization ability through the principle of structural risk  minimization.It is mostly used to solve the classification problem and regression problem of small samples.However,when used for prediction,a single model has certain limitations.In this paper,an improved multi-dimensional gray model and a support vector machine combined forecasting model are proposed.The combined forecasting model realizes the complementary advantages of different models,and can avoid the limitations of  the single model,increase the stability of the model.The experimental simulation results show that the proposed combined forecast.The prediction effect of the model is significantly better than the support vector machine and the innovation-based prioritization method,the prediction accuracy of the combined prediction  model is higher than that of the single prediction model. 

Key words: Forecasting model, Multidimensional grey model, Support vector machine, The combined forecasting model