A Nearest Neighbor-Radial Basis Function Network to Forecasting Grain Production

摘要

It is difficult to forecast complex nonlinear time series accurately by traditional prediction methods. A novel time series forecasting method named nearest neighbor-radial basis function networks (NN-RBFN) was designed in this paper through integrating nearest neighbor method with radial basis function networks. The algorithm of NN-RBFN is simple and easy to program, and it can be applied universally. NN-RBFN was applied to forecast our country's grain production, and satisfying results were obtained. The forecasting results are helpful to government decision-making.