Market Risk Measurement And Precaution Of Stock Index Futures Based On Assets Structure Under High-frequency Environment

Driven by continuous financial innovation,financial derivatives,such as stock index futures,open an important time window for development.Although regarded as the most effective risk management tool in financial markets,stock index futures has not been entirely safe.The market risk of stock index futures may even be far beyond the spots under the impact of the investor structure dominated by speculators and hedgers.Since the global financial turmoil caused by the US subprime crisis,the market risk of financial derivatives,including stock index futures,has been the focus of academic and industry attention.In April of 2010,the CSI 300 index futures were officially listed as the first stock index futures in China,creating a history of multi-tier financial markets.As one of the tough battles to forestall and defuse financial systemic risks,to reasonably define the market risk of stock index futures,scientifically measure and forestall the market risk of Chinese stock index futures,fully maintain the safe and smooth operation of the stock index futures market,then and finally realize high-quality growth of financial markets and the whole national economy,which is an important subject to be solved urgently in the field of financial engineering and risk management.With the development of information technology and digital storage technology,high-frequency time series analysis has been a more precise technical direction for the research of market risk of stock index futures.Making high-frequency realized volatility(RV)methods as the breakthrough point,this paper empirically measures and describes market risk of stock index futures,and provides countermeasures and policies to prevent market risk of stock index futures.Taking into account the two different situations in which the assets structure of investors is held by holding the futures and both the futures and spots,market risk of stock index futures is defined,and the sources and causes of market risk are identified,where stock index futures is expressed as “the probability of the potential loss of the investors' assets by the deviation from the expected value of future price volatility of stock index futures or the portfolio of the futures and spots”,and market risk of stock index futures are measured under the conditions of single futures and the portfolio of the futures and spots.Under the condition of single futures,firstly,the classical RV,realized range-based volatility(RRV),and realized bipower volatility(RBV)in high-frequency RV methods are respectively constructed to investigate the five-minute high-frequency price volatility of CSI 300 index futures.It is found that RBV has the best performance,while the weighted RRV(WRRV)is the second.High-frequency RV methods could well describe the discontinuity and jump in the price volatility of CSI 300 index futures;Then,market risk of stock index futures under the condition of single futures is measured respectively by WRRV-VaR and RBV-VaR model,and the results are compared.Compared with normal distribution and t location-scale distribution,the logarithmic normal distribution can better fit the peak and unilateral characteristics of WRRV and RBV.The estimates of Monte Carlo simulation show that both WRRV-VaR and RBV-VaR model can describe the market risk characteristics of stock index futures to some extent,and RBV-VaR model can be better in extreme situations.Under the condition of portfolio,in order to capture the nonlinearity,asymmetry and dynamic change of the dependence structure between high-frequency prices of CSI 300 index futures and spots,firstly,high-frequency RV method is adopted to fit the marginal distributions of time-varying Gumbel Copula,time-varying Clayton Copula and their mixture Copula,in which the dependence in the bear markets shows slightly stronger than in the bull markets.The goodness of fit tests show that the high-frequency RV-time-varying mixture Copula function has better measure ability;Then,this Copula function above is combined with the VaR method to measure market risk of stock index futures under the dynamic change of the investment proportion of the futures and spots.The estimates of Monte Carlo simulation show higher market risk of stock index futures with more investment in futures at the same confidence level and higher confidence level in the same investment proportion,while there is stronger linkage effect between stock index futures and spots in extreme situations.The model shows great performances all in 90%,95% and 99% confidence level.Based on the measurement of market risk and the different assets structure above,countermeasures for the market risk precaution,where the optimal hedging ratio strategies of the investors and margin level of the exchange are calculated under the condition of market risk minimization.Finally,policy suggestions for the market risk macro-precaution are put forward based on the analysis of the theory foundation,international experience and basic status of market risk precaution of stock index futures,as the application of theory and empirical research results.