A Study Of The Correlation Between Major Assets Based On The DCC-MVGARCH Model

With the development of China's financial market,the size of institutional investors is continuously expanding,and there are more and more varieties of financial assets that can be invested in the market.To reduce risks and increase investment returns,it is particularly important to carry out reasonable and effective asset allocation;financial market as a whole,studying the correlation between different types of assets and their dynamic correlations is a prerequisite for effective asset allocation.At the same time,as China accelerates its opening up of financial markets,following the introduction of Shanghai Stock Connect and Shenzhen Stock Connect,A shares will soon be incorporated into the MSCI Emerging Market Index.These measures will make China's financial market and the international market gradually converge,and the links between the markets will become closer.Therefore,it is of great practical significance and theoretical guidance to study the correlation between domestic large-scale assets and the linkage between China-US stock and bond markets.This thesis empirically analyzes the correlations and dynamic correlations between major domestic assets and China-US stock and bond markets.After using the common empirical analysis methods: cointegration test method,Granger causality test method,and variance decomposition method to analyze the static correlations among various assets,then the DCC-MVGARCH model was introduced to demonstrate the dynamic correlations among assets.This thesis gets the following conclusions:Relevant aspects of domestic large-scale assets:(1)Most of the fluctuations in various market returns are due to the market itself.Changes in stocks,bonds,and gold yields are greatly affected by fluctuations in commodity yields.(2)There is a two-way Granger causality between the stock market and the bond market,and there is no Granger causality relationship with other three types of assets.There is a two-way Granger causality between bonds and commodities,and the one-way Granger causality relationship with foreign exchange.There is also a unidirectional Granger causality between commodities and foreign exchange,gold and foreign exchange.(3)Dynamic correlation: there is a weak and unstable dynamic correlation between stocks and bonds,and there is a negative correlation during most of the time,but there are also positive correlations in a considerable number of periods,and the correlation coefficient is converted several times.Most of the time,the dynamic correlation between stocks and commodities is a strong positive correlation.When the financial market turmoil is large,it may turn into a negative correlation.There is a long-term,stable,weakly negatively correlated dynamic relationship between bonds and commodities,with only a small number of outliers being weakly positively correlated.In addition,dynamic correlations among other assets are weak.Relevance of China-US stock and bond markets:(1)There is a one-way Granger causality between the China-US stock and bond markets: changes in the US market may cause changes in the Chinese market,whereas changes in the Chinese market will not cause changes in the US market.It can be seen that the US financial market has a greater impact on China's financial market,while China has little influence on the US market.(2)The dynamic correlation between the stock market of China and the US is a strong positive correlation for most of the time.During the financial crisis and the period when the Chinese stock market turmoil is relatively large,the dynamic correlation coefficient between the two is more volatile.After the US financial crisis,the linkage between the stock markets of China and the US also increased.When the financial market turmoil is relatively large,the impact of the US stock market on the Chinese stock market is greater.The results of the dynamic correlation between China and the US bond market show that there is a weak positive correlation between the two markets,and only a few abnormal values are shown as negative,and the volatility is small.It can be seen that the US bond market has less influence on China's bond market,and the two are still relatively divided.Finally,based on the research conclusions,relevant suggestions are put forward: investors should make full use of the principle of correlation between assets when deploying large-scale assets,and also pay attention to the linkage of international financial markets;regulators should encourage the development of institutional investors and increase domestic while opening up financial markets,and strengthen the risk prevention of financial markets.