Research On The Co-movement Of Major International Stock Markets Before And After The Stock Market Crash

Under the background of global financialization and economic integration,the capital of a country can flow in other countries.That is to say,the fluctuation of any financial market may affect or even infect the financial market of other countries.With the rapid development of China's economy and the acceleration of the process of opening up to the outside world,the interdependence between China's financial market and other countries' financial markets is increasing closely.In recent decades,financial crises have occurred frequently,and the mutual dependence of financial markets makes the crisis spread to other financial markets,which attracts wide attention of scholars.Therefore,the study of the correlation and changes of stock markets among different countries during the financial crisis has certain practical guiding significance for investors to prevent financial risks and relevant departments to formulate macro-financial policies.This article selects three major stock markets in Asia,Europe and America to analyzed,including the Shanghai Composite Index,the Hang Seng index,the Nikkei 225 index,the KOSPI index,the Frankfurt DAX index,the FTSE 100 index,the Russia's RTS index,the Standard & Poor's 500 index,the Toronto stock exchange composite index and the MXX index in Mexico.On the basis of constructing marginal distribution model to filter yield sequence characteristics,the DCC-GARCH model was established to study the temporal variability,the Copula model was established to study its dependent structure,and the wave conduction mechanism was studied through VAR model.Firstly,descriptive statistical analysis of the data is conducted,and then the stationary test,normal test,autocorrelation test and ARCH effect test are carried out on the data.The results show that the data are non-normal and stable,with sharp peak,thick tail and ARCH effect.Therefore,the EGARCH-t model is established to filter the characteristics of the original yield sequence,and then the POT model is adopted to model the marginal distribution.Then,the DCC-GARCH model of Shanghai stock market and other countries(regions)is established,and the results showed that the time-varying correlation coefficients of Shanghai stock market and other countries(regions)declined but then quickly recovered in most cases in 2015.The standardized residual of EGARCH-t model was extracted,and the vine copula model was established for the whole sample interval,data before and after the stock market crash.The results showed that the Asian stock market,European stock market and American stock market showed regional aggregation.After the crash,the link between Asian and European markets was Hong Kong and the UK,instead of Hong Kong and Russia.The link between European and American stock markets was Germany and the United States,instead of Britain and the United States.Finally,the wave conduction mechanism of countries with dependency is analyzed.That is to say,a binary vector autoregression model is established for the stock markets of all countries(regions)with dependency before and after the stock market crash,and the Granger causality is tested.The results show that before the stock market crash,granger causality between stock markets in many countries(regions)is not significant.After the stock market crash,the lag order of the VAR model becomes bigger and the model becomes more complex,which may be caused by the stock market turbulence.Meanwhile,granger causality between stock markets of many countries(regions)is significant,indicating that the stock market of various countries(regions)becomes more connected and the volatility transmission relationship becomes stronger due to the stock market crash.