Oil Prices Forecasting Using Modified Support Vector Machines

Oil is a kind of basis energy, its price fluctuations have an important impact on the operation of the world economy. AS the non-linear features of world oil prices, the paper uses support vector machines(SVM) technology for the oil price forecast. The method can be effective in the data space of the evolution operating of various non-linear to the corresponding linear operation in characteristics space, thereby greatly enhancing its ability to handle non-linear. To solve the problems of SVM in training for large-scale convergence, such as slow convergence, greet complexity, particle swarm optimization(PSO) is proposed for the secondary planning problem to enhance SVM computing speed. The modified SVM is applied to oil prices forecast, empirical studies show that the method has a high prediction accuracy and faster computing speed.