Research On The Relationship Between China's CSI 300 Index And CSI 300 Futures Based On High-frequency Volatility

Since 1982 the United States lauched stock index futures, the trading of stock index futures in the global market is rapidly unfolded. Meanwhile, the research of the relationship between stock index futures and spot market has become a focus of attention of scholars. This paper based on the domestic and foreign research of the impact of stock index futures on the spot market, innovatively uses high-frequency volatility as the main line, firstly discusses the impact of stock index futures on stock market volatility is whether increasing or decreasing, secondly examines the real connection of stock index futures volatility and stock market volatility after the introduction of stock index futures, and at last does in-depth study on whether the causal relationship is caused by the continuous part or the discrete part of the volatility.When we discuss the impact of stock index futures on stock market volatility, we firstly use realized volatility and realized bipower variation of CSI 300 to measure high-frequency fluctuations, and secondly adopt the MEM model to fit the volatility sequence before and after the introduction of stock index futures. With the above two volatility sequence, we successfully prove that the introduction of stock index futures can reduce the volatility of the spot market, which is statistically valid in our paper.In chapter 3, we not only take advantage of traditional linear Granger causality test, but also innovatively consider the application of nonlinear Granger causality test, as we study the real connection of stock index futures volatility and stock market volatility. As a result, from the perspective of the high-frequency volatility, the linear and nonlinear Granger causality test both prove that CSI 300 index volatility one-way guide stock index futures volatility, which is similar to many emerging markets.Chapter 4 firstly introduces MinRV and MedRV which are two new volatility estimator, then does simulation of MinRV, MedRV together with RBV and RTV in order to select an optimal estimator that is closer to the real volatility. After extensive simulation, we find that there is little difference between the MSE of four above estimators, however, RBV is closest to the real volatility. Finally in chapter 5, this paper concludes that the causal relationship between stock index futures volatility and spot market volatility is mainly caused by the continuous part volatility by means of linear Granger causality test, meanwhile, nonlinear Granger causality test proves that the continuous part and the discrete part both lead to the above causal relationship.