Forecasting Short-Term Corn Price Changes: Using Artificial Neural Networks

The price expectation of agricultural products has been a thorny problem and many scholars have done research on this issue. However, network system is seldom applied in the research process. So, in order to obtain a more accurate method of forecasting the grain price, this thesis adopts the latest intelligent predicted method of neural network and establishes a modified neural network model. Through studying the fluctuation and growing trends of the grain price time-series data, this model predicts the future price by means of its self-learning characteristics in the end. The results show that: Since the neural network model can well fit the nonlinear problem, it has a high accuracy in predicting the price. This method has not only feasibility but also applicability and the prediction is objective and reasonable. So it has important research value and good application prospect.