As China's market-oriented interest rate process continuously push forward,the fluctuation of market interest rate is becoming more and more frequent.Interest rate risk has become the main risk faced by Chinese national debt market.The government bond futures is an effective tool to manage interest rate risk.In September 2013,China launched five-year Treasury futures,followed by the launch of 10-year Treasury futures in 2015.It provides more options for investors to manage the interest rate risk of treasury bonds.When investors are hedging,they establish a certain proportion of futures positions that are opposite to their spot positions to hedge against the risk of spot price fluctuations.The ratio of the futures to spot is the hedging ratio,which is calculated by a certain measurement model is the key of the hedging of the Treasury bond futures.Firstly,the paper teased out the related literature of the determination method of the optimal hedging ratio at home and abroad systematically,according to whether the determined hedging ratio changes with time,the hedging model can be divided into static hedging model and dynamic hedging model.Then,four widely used optimal hedging ratio determination models are introduced in detail,which are OLS,ECM,GARCH and Copula-GARCH respectively.Next,the paper expounds the evaluation methods of hedging effect.On this basis,we have selected daily closing price of four Treasury index of different maturity periods which are 3-5 years,5-7 years,7-10 years,and more than 10 years respectively,CTD and 10-year Treasury futures as samples.The sample interval was from April 15 th 2015 to October 17 th 2017,in which 407 data from April 15 th 2015 to December 30 th 2016 were selected as in-sample data,and 177 data from Janurary 3rd 2017 to October 17 th 2017 were selected as out-of-sample data.Then we use the above four methods to estimate the optimal hedging ratio of each spot.Finally,in order to determine the best model of hedging and the spot which is most suitable for hedging interest rate risk with the futures of ten-year Treasury bonds,we use two methods which are risk minimization and unit risk return maximization to evaluate the hedging performance of empirical results from two angles.Then we also use the out-of-sample data to verify the applicability of the optimal hedging ratio to the future.The results show that if the risk minimization method is used to evaluate the performance,the hedging performance of the 10-year Treasury futures is best of all to the 7-10 year Treasury bond index,while the unit risk compensation maximization method is used to evaluate the performance,the 10-year Treasury futures have the best effect on the hedging of the 5-7 year Treasury bond index.For these five spot stocks,the OLS model is optimal by using the risk minimization method,and then we find the ability to avoid the risk of Copula-GARCH model is strongest with out-of-sample data.When we use the method of unit risk compensation maximization to evaluate the data both within and outside the sample,we can find the Copula-GARCH model is optimal all the time.Copula-GARCH model is a dynamic hedging model.The hedging ratio estimated by the modol of Copula-GARCH is changing over time.While the model of OLS which is simple and practical is a static hedging model.In the process of hedging,investors should choose the appropriate model to determine the optimal hedging ratio according to their attitude towards risk and income. |