基于改进加权移动平均法的铁矿石到岸价格预测

摘要

为有效预测铁矿石到岸价格的涨跌趋势及波动幅度，基于传统加权移动平均法在预测上的不足，引入趋势及趋势校正值的概念，提出一改进加权移动平均模型。以海关统计数据2009年1月—2014年5月铁矿石平均到岸价格对模型的可靠性进行验证。该模型分别对价格平稳期及价格波动期的到岸价进行短期预测，并预测2014年6月至12月的平均到岸价格。预测结果表明：采用改进的加权移动平均法在价格平稳期内且观察期间足够小时，预测值基本与实际值吻合。模型对短期内的价格变化趋势可以有效预测，但长期价格预测容易出现偏差，需要根据实际数据调整趋势值，以获得更准确的趋势判断。%To predict the ups and downs of trend and the fluctuation range of iron ore CIF price effective-ly,the concepts of trend and trend correction are introduced and an improved Weighted Moving Average (WMA)model is proposed considering the defect of the traditional WMA method in forecasting. The reli-ability of the model is verified by the customs statistical data of iron ore average CIF price from Jan. 2009 to May 2014. The model separately predicts the short-term CIF price in stable and fluctuant periods. The average CIF price from Jun. to Dec. in 2014 is also predicted. The results show that the predicted value obtained by the improved method is in line with the actual value in the stable period when the observation period is sufficiently small. The short-term price trend can be effectively predicted with this model,but the long-term price forecasting is prone to error. So it is necessary to adjust the trend according to actual data in order to obtain more accurate judgment on price trend.