Study On The Jump Phenomenon Of Chinese Stock Market

Researching in the stock return volatility time series of jumping phenomenon is one of the focus in recent years,which is mainly because that jumping phenomenon is wide-spread in various fields of financial economic,hasing important significance.At the same time,in recent years benefiting from the using of high-frequency data,empirical studies of this phenomenon can be further developed.A lot of empirical evidence show that,by adding the jump component to the financial asset price modeling process,the model fitting degree for yield,volatility and other indicators increased dramatically;and for jumping phenomenon itself it's also a very good study reveals yields characteristics of non-continuous changes in the sequence.This makes the research of jumping phenomenon plays an important role in a number of aspects such as risk hedging,product pricing.In this paper,based on the theory of market microstructure,using the BTL method,we study the China CSI 300 index which plays a role as a "barometer" of the stock market of Shanghai and Shenzhen.We analyze the causes of jump and jump in volatility forecasting and risk prediction.The main work of this paper is as follows:First we choose the data of China CSI 300 index from January 4,2007 to December 30,2013 as research sample and then using the widely used BTL model for the recognition and separation of jump.After this we firstly give a statistical description about the result.Then in consideration of the close relationship between the stock market and the Macroeconomic situation of China,we analyze the relationship between the jump and releasing of the macroeconomic news by using the Logit model.Then we optimize the HAR-RV model by taking the jump phenomenon into consideration,which improves the forecast results of Realized-Variation.At last we use the optimized model to forecast the VAR and compare the result with the traditional GARCH model which indicates the important role of the non-continuous components in the market,that is,jump to the volatility of returns and risk prediction.