The Research On Realized Volatility In China Stock Market

For the last 20 years, volatility has been the hotspot of the financial economics. Since the first conditional volatility model by Engle(1982), thousands of papers concerning conditional heteroskedasticity have been published. Most recently, Anderson, Bollerslev, Diebold,Labys and other economists developed a new estimator of volatility——"realized volatility".Although some economists have developed and investigated the measuring methods of realized volatility, there remain some deficiencies. First of all, asset price series does not follow normal diffusion process exactly, so the frequency of date used to estimate volatility should not be too high. Date of very high frequency would bring more errors due to microstructure friction.Torben G.. Andersen and his coworker chose a date interval of 5 minute when studying the volatility of Dow Jone 30 stocks. However the optimum data interval to balance the usual measuring error and the microstructure error would not be unique along all the sample periods. It seems to vary in different periods and different markets.This paper use the model of Bandi and Russell to give the frequency of realized volatility, the result is 13 minutes in China stock market, and I also test the distribution of realized volatility,we find that the distribution of realized volatility is highly right-skewed,while the realized logarithmic standard deviaton is approximately Gaussian, as is the distribution of the returns scaled by realized standard deviation. We also test long memory properties, and find the d=0.2639.also, we test the relation between realized volatility and volume, the result support the mixture of distributions hypothesis. Finally we concluded the content of this paper and put forward several directions in future study.