考虑多因素的铁矿石价格线性回归预测

Linear Regression Prediction of iron ore prices considering multiple factors

铁矿石价格受多种因素影响,对其进行准确预测较为困难,尤其是在近些年铁矿石价格波动明显的情况下。结合1995—2015年间铁矿石价格统计数据,着重分析了对铁矿石价格影响较为明显的GDP、固定资产投资、钢材产量、废钢消耗量、原矿供给量以及铁精粉生产成本等六大因素,建立了多元线性回归预测模型,对未来五年铁矿石市场价格进行了科学预测,得到未来五年铁矿石市场价格分别为555.56元/t、613.94元/t、665.90元/t、709.93元/t、744.35元/t,年均增长率约为5.70%。研究表明,2016年的预测值与该年度市场实际价格较为接近。预测值可在一定程度上反映未来五年铁矿石市场价格变化趋势,为相关企业决策提供参考。

Iron ore prices are affected by a variety of factors,and it is difficult to predict it accurately,especially in the case of significant volatility in recent years.Combined with the statistical data from 1995 to 2015,analyze six factors which impact iron ore prices significantly including GDP,fixed investments,steel production,consumption of scrap steel,raw ore quantity supplied and cost of production,then establish multiple linear regression prediction model.So the prices of iron ore in the next five years are predicted,which are 555.56、613.94、665.90、709.93and744.35 RMB Yuan/ton,respectively,and the average annual growth rate is about 5.70%.The predicted value in2016 is close to the actual market prices of the year.The predicted value will reflect the trend of the iron ore market prices in the next five years,which can provide reference for the related enterprise to make decision.