The Application of Support Vector Machine in Forecasting Grains Futures

In contemporary finance, derivatives such as futures and options have exerted increasingly prominent impact not only on risk management activities but also on price speculative activities. As a result of the high-leverage characteristic involved in derivative tradings, investors can gain enormous profits with a small amount of capital if they accurately forecast the market trend. The support vector machines (SVMs) have been successfully applied in solving nonlinear regression and times series problems. In this study, the SVMs are employed to examine the feasibility in forecasting grain futures. In addition, two other forecasting approaches, namely the autoregressive integrated moving average (ARIMA) model and general regression neural network (GRNN) model, are used to compare the performance of forecasting. Experimental results show SVM model is a valid and promising alternative in grain future forecasting