Domestic And International Correlation Study On Volatility Of Shanghai Crude Oil Futures

In the past,financial asset research using daily yields often neglected the fluctuations information implied by high-frequency intra-day data.Therefore,this paper uses the mixed data which combines daily yield and intra-day yield and multiple Copula functions,and takes China's newly introduced Shanghai crude oil futures as the research object.It studies quantitatively the static correlation and dynamic correlation between Shanghai crude oil futures and domestic and international crude oil futures and spot.At the same time,it compares and analyse the correlation coefficient between Shanghai crude oil futures and other crude oil futures and spot.This paper selects the 30-minute closing price and the daily closing price of the continuous contract of each crude oil futures and daily closing price of crude oil spot from March 26th,2018 to December 28th,2018.After that,the previous correlation results are applied to the estimation of Shanghai crude oil futures VaR and CVaR,,and the accuracy of the model estimation results is tested.The empirical results show that the correlation between Shanghai crude oil futures and international major crude oil futures are weak both statically and dynamically.This may be due to the fact that Shanghai crude oil futures market has not been run for a long time,and foreign investors have insufficient participation.The Shanghai crude oil futures market is not sufficently open to the outside world and crude oil varieties of the delivery target are different.However,it has a certain correlation with the correlation between Shengli crude oil spot and downstream product asphalt futures.From the prospective of dynamic correlation,the dynamic correlation coefficient between Shanghai crude oil futures and other crude oil futures changes around the static correlation coeff-icient.When a risk event occurs in the international crude oil market,the correlation between crude oil futures markets will be strengthened in a short period of time.Finally,the correlation results of Shanghai crude oil futures and asphalt futures are applied to the VaR(CVaR)estimates,comparing the traditional methods with the VaR(CVaR)estimated by the GARCH-MIDAS-Copula model.At different confidence levels,the VaR(CVaR)obtained by the GARCH-MIDAS-Copula model exceeded the estimates of the traditional method.At the same time,a backtesting test was carried out on the estimated results of the model.