摘要
本文首次对影响中国原油期货价格波动的驱动因素进行了量化分析.运用广义动态因子模型,结合互联网数据,为中国原油期货价格构造了六类预测因子:供需预测因子、市场金融化预测因子、汇率市场信息预测因子、商品市场预测因子、全球宏观经济预测因子以及事件预测因子.基于混频GARCH-MIDAS模型,本文发现上述六类因子能显著改善对原油价格波动的预测精度.同时,基于MCS检验结果,揭示出在不同时间尺度下,驱动中国原油价格波动的信息存在明显差异性,即在短期和中期预测中事件预测因子起主导作用,而供需因子则是在长期主导中国原油价格波动的关键因素.综上,本研究为国内原油市场参与者、政策制定者及市场监管者把握未来市场信息提供了分析工具和参考依据.
This paper quantifies for the first time the driving factors predicting price volatility of China’s crude oil futures.Using the generalized dynamic factor model,and the internet data,this paper constructs six factors to predict the price movements of China’s crude oil futures:the demand-supply factor,market financialization factor,exchange rate factor,commodity market factor,global macroeconomic factor,and event-driven factor.Results from the GARCH-MIDAS model show that these factors can significantly improve the forecasting accu-racy of crude oil futures price volatility.Further,results based on the MCS tests demonstrate that the underly-ing information driving price volatility varies across different time horizons.Specifically,the event-driven fac-tor plays the leading role in the short-term and medium-term forecasting,whereas the demand-supply factor is the key for explaining long-term price volatility.Overall,this study provides a useful framework and key refer-ence for market participants,policymakers and regulators to utilize market information in China’s crude oil market.
作者
马嫣然
吴菲
张大永
姬强
MA Yan-ran;WU Fei;ZHANG Da-yong;JI Qiang(Institutes of Science and Development,Chinese Academy of Sciences,Beijing 100190,China;Research Institute of Economics and Management,Southwestern University of Finance and Economics,Chengdu 610074,China)
出处
《管理科学学报》
CSSCI
CSCD
北大核心
2024年第1期113-125,共13页
Journal of Management Sciences in China
基金
国家社会科学基金资助重大项目(23&ZD093)
国家自然科学基金资助项目(71974181,71974159,72303219)。
作者简介
通讯作者:姬强(1982-),男,山东威海人,博士,研究员,博士生导师.Email:jqwxnjq@163.com。
