摘要
中国铁矿石价格指数(CIOPI)自2011年10月起公开发布,这不仅对铁矿石买卖双方具有重要的价格指导意义,更是我国在争取铁矿石定价权上取得的实质性进展.为探究中国铁矿石价格指数的内在规律,以2017年5月8日至2018年6月1日的数据为训练集,用Holt-Winters滤波和ARIMA模型分别进行拟合和预测,对比未来5天的预测值与实际值,发现ARIMA(3,1,3)模型预测效果最好,平均相对误差为1.09%.结果显示,未来5日中国铁矿石价格指数将在235附近波动,呈现小幅上涨趋势.
The China Iron Ore Price Index(CIOPI)has been published since October 2011,which has the meaning of price guidance for both iron ore buyers and sellers.It is also the substantial progress we have made in securing iron ore pricing power.In order to explore the inherent law of China’s iron ore price index,this paper takes the data from May 8,2017 to June 1,2018 as training set,and uses Holt-Winter method and ARIMA model to fit and forecast respectively.By comparing the predicted value with the actual value in the next five days,it is found that ARIMA(3,1,3)model has the best prediction effect,with an average relative error of 1.09%.The results show that China’s iron ore price index will fluctuate around 235 in the future,showing a slight upward trend.
作者
杨锦伟
李宜懋
YANG Jin-wei;LI Yi-mao(Institute of Mathematics and Statistics,Pingdingshan University,Pingdingshan 467000,China;Department of Statistics,Henan University of Economics and Law,Zhengzhou 450002,China)
出处
《数学的实践与认识》
北大核心
2020年第11期289-298,共10页
Mathematics in Practice and Theory
基金
河南省科技厅软科学项目(192400410074)
平顶山学院博士启动基金(PXY-BSQD-2019005)
平顶山学院应用型课程建设项目(2019-YYXKC14)。
