大宗矿产品价格预测方法的发展综述  

Review on the development of price forecasting methods for bulk mineral products

作　　者：潘昭帅 张照志[1] 吴晴 PAN Zhaoshuai;ZHANG Zhaozhi;WU Qing(Ministry of Natural Resources,Key Laboratory of Metallogeny and Mineral Assessment,Institute of Mineral Resources,Chinese Academy of Geological Sciences,Beijing 100037,China;China University of Geosciences(Beijing),Beijing 100083,China)

机构地区：[1]中国地质科学院矿产资源研究所,自然资源部成矿作用与资源评价重点实验室,北京100037 [2]中国地质大学(北京),北京100083

出　　处：《中国矿业》2022年第3期20-30,共11页China Mining Magazine

基　　金：中国地质调查局地质调查项目“全国矿产资源国情调查与综合评价”资助(编号:DD20190606)。

摘　　要：大宗矿产是国民经济建设中具有举足轻重地位的主体型矿产,准确预测大宗矿产品价格和趋势,对于保障国家资源能源安全,推动经济全球化具有重要意义。方法的选择是进行大宗矿产品价格预测的核心,而随着方法日益多样化、复杂化、学科交叉化,合理地对预测方法进行分类的难度不断加大。本文通过对相关文献进行分析,梳理了大宗矿产品价格预测方法的发展历史,在总结前人分类的基础上,根据预测方法发展的趋势和特点,提出了大宗矿产品价格预测方法分类的新思路。根据文献的总结研究,认为现有的预测方法可以按照预测精度、学科、技术和形成背景四个要素,将其分为定性预测法、计量经济预测法、人工智能预测法和混合模型预测法四类。通过对各类方法优缺点的分析对比认为,该分类可操作性强、应用范围广、与时俱进,能够更好地满足不同任务的需求,进而为提高方法选取效率、提升预测精度提供有效支持。综上,结合当前时代背景,提出了未来价格预测难度陡增、新分类适用性好、混合预测模型逐渐成为主流、正确认识方法模型的局限性等结论与展望。Bulk minerals are the main minerals in the national economic construction.It is of great important to accurately predict the price and trend of bulk mineral products,which is of great significance to ensure the national resource and energy security and promote economic globalization.The choice of methods is the core of price prediction of bulk mineral products.With the diversification,complexity and interdisciplinary of methods,the difficulty of classifying the prediction methods is increasing.Based on the analysis of relevant documents,this paper combs the development history of the price prediction methods of bulk mineral products.Based on the summary of previous classification,according to the development trend and characteristics of the forecasting methods,a new way of classification of the price prediction methods of bulk mineral products is put forward.According to the literature summary and research,the existing forecasting methods can be divided into four types:qualitative forecasting method,econometric forecasting method,artificial intelligence forecasting method and mixed model forecasting method according to four factors,including prediction accuracy,discipline,technology and formation background.Through the analysis and comparison of the advantages and disadvantages of various methods,the research show that the classification can be operated,widely used and kept pace with the development,and can better meet the needs of different tasks,and thus provide effective support for improving the efficiency of method selection and improving prediction accuracy.In conclusion,combined with the current background,the paper puts forward the conclusions and prospects of the steep increase of price forecasting difficulty in the future,good applicability of new classification,the gradual mainstream of mixed forecasting model and correct understanding of the limitations of the method model.

关 键 词：大宗矿产品 价格预测 预测方法 预测模型 能源安全 

分 类 号：F205[经济管理—国民经济] F407.1