Import iron ore price forecasting based on PSO-SVMs model

According to the nonlinear series characteristic of the price of imported iron ore, this paper proposes a support vector machines (SVMs) model for import iron ore price forecasting. But parameters of SVMs model are very difficult to determined, particle swarm optimization (PSO) algorithms are used to search these parameters and make sure the accuracy of SVMs model. Compared with autoregressive integrated moving average (ARIMA) model and BP Neural Networks, SVMs model has the highest prediction precision, and the results of SVMs model are more tally with the actual situation.