改进的支持向量机石油期货价格预测模型研究

Research on Petroleum Price Prediction Based on SVM

研究石油期货价格预测精确度问题。石油价格预测随机性很强且受到市场复杂变化条件影响,是曲型非线性问题。针对传统线性关系的价格预测模型对石油的价格预测准确度较低,提出了一种改进的支持向量机石油期货价格预测模型方法,采用石油期货价格序列的一阶差分作为SVM的输出,一阶差分的若干滞后值作为SVM的输入。同时采用一种新的滞后阶数寻优方法,将滞后阶数与其它模型参数一样看待,使用验证集中技术获得所有参数的最佳值。最后实验采集了纽约商品交易市场石油期货价格数据作为实验数据,仿真结果表明,改进的价格预测模型提高了石油价格预测的准确度,是一种有效使用的石油价格预测模型。

The oil futures price forecasting accuracy problems.Oil price forecast has been the research focus at home and abroad,the price forecasting model for the traditional method of forecasting oil prices and low accuracy,this paper proposes an improved support vector machine method of the oil futures price forecasting model,using oil futures price series as the first difference of the output of SVM,a number of lagged first difference value as the SVM input.At the same time using a new method of lag order of the optimization,the lag order and the other model parameters like the look,the use of authentication technologies focus on the best value of all parameters obtained.The experiments were collected from the New York Mercantile Exchange oil futures price data as the experimental data,simulation results show that the improved price forecasting model improves the prediction accuracy of the price of oil is an effective use of the oil price forecast model.