Entropy-Based Combining Prediction of Grey Time Series and Its Application

Unit crop yield prediction is an important and widely studied topic since it can have significant impact on macroeconomic regulation and agricultural structural readjustment both in national wide and local. The grey system theory and neural network have individually been applied to the prediction problem of various fields, and have shown good results. However, few studies have dealt with the integration of grey system theory and neural network for the unit crop yield prediction, though there is a great potential for useful applications in this area. In this paper, based on related concept of information entropy, which is applied to determine the weights of grey system forecasting model and RBF (radial basis function) neural network forecasting model, an entropy-based combining prediction model of unit crop yield time series is proposed. This study proposes a novel combination forecasting model for improving those traditional analyzing and forecasting models as following: with respective merits of both grey forecasting model and RBF neural network forecasting model, the conjunction forecasting model is a comprehensive reflection of both social production levels and environmental factors, and it is less risky in practice, as well as relative more intuitive and feasible.