Vegetables Price Forecasting In Hainan Province Based On Linear And Nonlinear Combination Model

With Hainan vegetables price from 2012 to 2015 as the research object, Back Propagation (BP) neural network forecasting model and Autoregressive Integrated Moving Average (ARIMA) forecasting model of Hainan vegetables price are set up. Based on above two models, linear combination and nonlinear combination forecasting model of vegetables price are established by linear programming method and BP neural network method. The results indicate that nonlinear combination model is superior to linear combination model, and linear combination model is superior to any single model, reflecting that combination model can make individual models complement each other's advantages to improve prediction accuracy.