Empirical Study On The Measurement Method Of China Stock Market Risk Based On A Class Of GARCH Models

This paper studies the application of VaR and CVaR methods for measuring stock market risk in the China stock market,focusing on the Shanghai and Shenzhen 300 index and some representative single stock from 2006 to 2015.The calculation results of several kinds of models are compared and their specific application conclusions are made.At the same time the empirical analysis on the current risk situation of China stock market is given.The accurate judgement of the stock market risk is not only investors and the theoretical researchers must pay attention to,but also the key to effective management and prevention of risk.VaR and CVaR methods in the measurement of stock market risk are still the focus of application research.Therefore,it's of practical value and theoretical significance to study the risk measurement of stock market.This paper first outlines the basic principles calculation methods of VaR and CVaR and then tests the stability and correlation of stock returns sequence using the VaR and CVaR methods based on the big data in China stock market.Next,the stock price volatility model is established,the risk value is calculated,and the results are tested.The empirical test has achieved good results.This paper uses statistical analysis softwares and the econometrics methods to analyze based on the abundant and accurate data.The empirical analysis shows that(1)the stock markets have "peak,fat tail" characteristics.(2)The results which are based on GED distribution are better than the results based on normal distribution and t distribution.(3)CVaR is accurate when VaR fails to measure the extreme losses.Compared with VaR,CVaR is a risk measure index that can cover a wider range of tail risks.