Research On Risk Measurement And Risk Conduction Of Chinese Stock Market

Since the1980s, the volatility of financial market is increasing, which greatly impact national economic and financial situation, affecting the healthy development of market economy and the financial markets all over the world. With the China accession to the WTO, China market gradually open to the outside world.While the international contact brings to the development of financial market Chinese opportunities, it also brings many risks and uncertainties. Governments and scholars have begun to realize the importance and necessity of financial risk research.They gradually began to issue the research in the field of financial risk.Quantile regression method is a semi-parametric method which does not require the setting of normal distribution assumptions and parameters for fat tail characteristics of financial time series and it has a very important significancein the study of risk measurement and risk aspects of conduction.This article will mainly use quantile regression conducted risk measurement and risk conduction from the perspective of China’s securities market theory and empirical research.Firstly, in the terms of the risk measure, in view of the unique situation of China’s securities market, there is a certain gap in the measurement of VaR (value at risk) metrics between our country and abroad.This paper puts forward an innovative measure of liquidity adjusted VaR levels, under Engle in CAViaR model proposed recursive quantile regression method which considers the size of liquidity will have an impact for the future of the level of risk.So on the basis of existing CAViaR model liquidity indicator variables are introduced,then I propose liquidity adjustment CAViaR model and established a VaR backtesting inspection framework. The results show that China’s stock market liquidity risk of future changes have a significant impact.Compared to the indirect GARCH models,the improved models’ out of sample forecasting performance are better than the indirect GARCH models’out of sample forecasting performance.And the predicted results of the improved models also have significant elevation. The liquidity adjustment CAViaR model is better than the indirect GARCH model in the characterization of the evolution of the market risk model.Secondly, in the terms of the risk conduction, unlike previous scholars who conducting research on risk is limited to unilateral qualitative analysis or risk spillover measure of quantitative analysis, this article is a combination of risk-Granger causality test and quantile regression methods for Shanghai and Hong Kong markets risk conduction qualitative and quantitative analysis. In this paper, the CSI and Hong Kong market index data are used for the emperical research.The first step is to estimate the best VaR level of risk based on the existing CAViaR models.Then use the risk-Granger causality test to verify the relationship between Hong Kong and Shanghai and Shenzhen stock markets, followed by the use of Co VaR methods to measure the risk of spillover strength between the various markets and stages are compared. The results show that there is a risk conduction relationship between Shanghai, Shenzhen and Hong Kong stock markets and there is a positive two-way risk spillover between Shanghai, Shenzhen and Hong Kong stock markets.And risk spillover between Shanghai and Shenzhen market was stronger than the spillover effects between the Hong Kong stock market and the other two.Compare the two phases spillover effects, risk spillover between various markets is gradually become stronger.In this paper, two aspects of risk will be the quantile regression method is applied to measure and China securities market risk conduction, using CAViaR model and CoVaR model to quantile regression as the basic thought of the study, obtained the satisfactory conclusion, studies show that the application of risk based on the quantile regression method is applied to Chinese securities market.