Research On Volatility Of China's Stock Market Based On Nonparametric Estimation Method

We all know that the stock market in our country started relatively late,which opportunity and risk can coexist.China stock market is very sensitive to external stimuli,and its volatility will reflect on the stock price.There are several reasons: first of all,with the rapid development of information technology and the globalization tendency of the money market,new financial products continually come forth so that the complexity of financial investment increased.Secondly,the return distribution has a peak and a fat tail.The stock market has the same feature of non-symmetry like the mature markets.Moreover,our stock market never got rid of the fact of "Policy Market",the stock price is unstable due to the policy changes.It shows that people should strengthen the study of financial steady and control money market.The multi-factors of economic activities,stochastic volatility,the irreversibility of events,and the non-stationary nature of time series have always affected the scientific process of economics.Economic problems have no functional relationship,only statistical models can be established.So now the non-parametric estimation method begins to glow in the financial market.Based on the highest and lowest prices of SI,ZI,HS300 and HSI,and daily range data of SI,ZI,HS300 and HSI.This paper Combines classical parameter models,such as GARCH model and CARR model,and non-parametric methods,the volatility of financial markets is studied and analyzed.The conclusion is that the wave function obtained by the nonparametric regression method is not constant.When the stock return rate rises,the volatility also increases.In the stock market,the positive impact is greater than the negative impact.The non-parametric GARCH model can be used as a complement to the parameter GARCH model without considering the model form and the yield distribution function constraints.The application of these methods also shows that various studies of nonparametric estimation of financial sequences have become an important part of financial econometrics.