Measuring Risk In Chinese Stock Market Based On GCARR-POT-EVT Model

In the context of the global economic infiltration,the economic development among countries is influenced by each other,and at the same time,it will inevitably face the impact of volatility,especially the volatility of the international financial market,such as the Asian financial turmoil in Southeast Asia in 1997 and the United States subprime crisis in 2008.The scale of the financial crisis leads to the more severe risk of the stock market in China,which makes the Shanghai Composite Index and the Shenzhen stock index fluctuate violently.On the one hand,this volatility increases the risk of the stock market to a certain extent,on the other hand,it is not conducive to the correct decision of the regulators and the broad investors in the stock market of China.The main reason is that the supervisors of the stock market in China should be able to measure the volatility and risk of the stock market in the extreme circumstances.Therefore,can we measure the risk of the Chinese stock market scientifically and control the risk of the stock market in the extreme case,and then put forward the corresponding measures to the regulators and investors in the stock market of our country.The risk aversion proposal has important theoretical and practical significance.Chou(2005)proposed the CARR(Conditional Auto-Regressive Range,CARR)model,this model is a combination of range and GARCH(Generalized Auto-Regressive Conditional Heteroskedasticity)modeling method of the model,but different from yield as the modeling object of GARCH model,CARR model is the effect on volatility modeling,to scientifically measure the extreme risk and volatility of the stock market in our country.There are many ways to measure the risk of stock market.Among them,Value at risk(VaR)risk measurement tool is highly recognized for its scientificity and accuracy in risk measurement and prediction.Extreme Value Theory(Extreme Value,EVT)are widely used in the 90 s in the field of risk measure,studies have shown that using Extreme Value method to evaluate risk,than the traditional method is more suitable for thick tail distribution data,and the predicted results are stable.An important model in extremum theory is called the POT(Peak Over Threshold)method,also known as the superthreshold model,which is the method of modeling all data in the sample that exceeds a given Threshold.Extension,this article through to the CARR Model on the Shanghai composite index and Shenzhen component index,data for empirical research,choose the empirical analysis of the data fitting the best of random perturbation terms With Gamma DistributionGCARR(Conditional Auto-Regressive Range Model With Gamma Distribution)Model,combined With the extreme value theory and its POT Model,build GCARR-POT-EVT Model for VaR risk measure of Chinese stock market.The specific ideas are as follows:firstly,the background,significance,research ideas and methods of research questions are elaborated,and the research trends in related fields are analyzed.Then,the paper reviews the CARR model,EVT and the risk measurement of Chinese stock market.Furthermore,the CARR model and its extension are introduced,and the GCARR-POT-EVT model is constructed.Then,the daily data of Shanghai composite index and Shenzhen composite index are selected respectively from 2009 to 2018,and the risk of China's stock market is measured by the VaR method.Finally,from the perspective of regulators and investors,the policy Suggestions on how to manage and avoid the risk of Chinese stock market are put forward.The main conclusion: GCARR-POT-EVT model is constructed,in the form of the GCARR model is estimated,and calculated the GCARR-POT-EVT VaR calculation formula of the model,the empirical results show that this article build GCARR-POT-EVT model than traditional GARCH class models and traditional CARR model can more accurately and effectively measure extreme stock market risk in our country.