Abstract:The paper firstly chooses grain reserves, grain production cost, grain production, grain policy, production demand, trade demand, and psychological expectations as seven main factors influencing grain price volatility based on the Lasso method. And then, according to Lasso variable selection, we conduct the regression and forecasting of grain prices with the help of support vector machine (SVM). At the same time, by comparing the fitting prediction effect of the Lasso, support vector machine (SVM), Lasso-SVM and ARIMA method, the empirical results show that the Lasso-SVM of fitting forecasting effect considerably overmatches the other three methods. |