The Application Of Volatility Factor Variable In Stock Index Futures’s High Frequency Trading Model

Chinese stock index market begin on april16,2010.With the launch of stock indexfutures,stock index futures market and stock index market have been the most important partsof financial investment market, and also have good influence on finacial market.Stock indexfutures has advantage of hedging、risk aversion、price discovery.then more and more investorsfocus on it,but it is also decided by many other factores,such as this article study the volatilityas the factor and how does the volatility affect stock index futures market.First,In this paper,we choose stock index futures market and stock index market asresearch target,and ues the five minutes data of CSI300stock index、CSI300stock indexfutures from Aprial19,2010to Feb20,2013.though using sas software to analysis thestatistical characteristic and distrubition of the data.,we get the results that the data of CSI300stock index futures is stable comparing with CSI300stock index,but it does not obey thenormal distribution and has the charactic of rush fat-tailed.Second,we use three different ways to define volatility,they are Garch model、Garman-krass eqution and SRV model.we render these volatility as factors which can affectfutures reward,we found that not only the volatility of CSI300stock index future but alsovolatility of CSI300stock index and stock index can impact on the return of CSI300stockindex future,that is to say volatility and stock index is the granger cause of stock indexfuture,though the cointergration test we also find casual relationship between them is long.Last,we set up9AR-regression model, the dependent variable is the return od CSI300stock index future,the independent variable are different volatility and CSI300stock indexreward.thongh comparing MSE and fitting effect,we found that the Garch3model is betterthan any others.This model describes the relation between volatility of stock index future、thecross term of volatility、CSI300stock index future and CSI300stock index reward.Whenstock index’s lag return change an unit,it will cause0.4unit on stock index future,in additionthat volatility and its cross term can also affect stock index future reward signally.we alsoprove that AR-RR model has good extrapolation.