Study On The Volatility Of Copper Future Returns Based On GARCH-MIDAS Model

China's consumption of refined copper has ranked first in the world for many years.In recent years,consumption has been close to half of the global consumption.Therefore,copper is crucial for China's construction and development.In this context,this paper combines the low-frequency economic variables with the high-frequency GARCH model to construct a generalized autoregressive conditional heteroskedasticity mixed data sampling(GARCH-MIDAS)model based on the mixed data sampling(MIDAS)method to analyze the impact of the monthly low-frequency supply and demand status and the macro-manufacturing status on the daily high-frequency return volatility of copper futures in China.The study finds that: First,the supply and demand states of refined copper market have a significant impact on the volatility of China's copper market return fluctuations.If the supply of refined copper relative to the demand increases,it will rise the volatility of copper futures returns,if the supply of refined copper relative to the demand decreases,it will reduce the volatility of copper futures returns;the increase of purchasing managers' index,that is to say,the increase in the prosperity of the manufacturing industry,will reduce the volatility of China's copper futures returns;the decrease of purchasing managers' index,that is to say,the decrease in the prosperity of the manufacturing industry,will rise the volatility of China's copper futures returns.The increase of the producer price index chain relative ratio will reduce the volatility of China's copper futures returns.The decrease of producer price index chain relative ratio will increase the volatility of China's copper futures returns.Second,investors and participants in China's copper futures market respond differently to good news and bad news.They respond more to bad news than they do when faced with good news,that is to say,they overreact when facing bad news.When investors and market participants face bad news,they will have a psychological state of loss aversion.At this time,the weight endowed by them to refrain from loss is much more than the weight endowed to benefit when faced with good news.So when the bad news in the current copper futures market occur,the volatility of the return of copper futures will be greater than the moment when the good news emerges.Third,The increase in the volatility of copper supply and demand will rise the volatility of copper futures returns.The decrease in the volatility of copper supply and demand will reduce the volatility of copper futures returns.However,the volatility in the macro state of manufacturing do not have a significant impact on the volatility of China's copper futures returns.Fourth,the GARCH-MIDAS model is used to forecast the volatility and value at risk in the Chinese copper futures market.The results show that: the generalization capability of GARCH-MIDAS out-of-sample forecast is excellent,and its prediction accuracy of the volatility and VaR is higher than that of the traditional GARCH(1,1).Based on empirical research results,this paper presents relevant policy recommendations from the perspectives of the national government,futures exchanges,futures investors,and futures companies to promote the healthy and orderly development of the copper futures market.