Research On Dynamic Correlation Of China’s Stock Market And Bond Market

The development of financial liberalization has made the links between financial sub-markets closer,and research on the correlation between financial markets has become an important topic.Since the 1990 s,China’s capital market has developed rapidly.The stock market and bond market are important components of the financial market..Grasping the trend of the correlation between stocks and debts helps investors optimize asset portfolio,manage risks and adjust hedge ratios to control risks in time.At the same time,it also helps financial regulators to regulate financial markets and regulate the macro economy better.As we all know,the correlation between financial assets presents dynamic and asymmetrical features.Therefore,by combining asymmetric DCC model and DCC-MIDAS model,this paper proposes a new asymmetric DCC with mixed data sampling model(Asymmetric Dynamic Conditional Correlation Model with Mixed Data Sampling,ADCC-MIDAS),and then portrays the correlation between China’s stock market and bond market.The ADCC-MIDAS model is a reasonable extension of the DCC-MIDAS model,which absorbs the characteristics of the DCC-MIDAS model and ADCC model.Its dynamic correlation can decompose long-term dynamic correlation components and short-term dynamic correlation components,and adds an asymmetric response to the positive and negative impacts of return on assets to capture the characteristic of asymmetric reactions.First of all,the data of this paper selects the Shanghai and Shenzhen 300 Index and the China Bond Bank Total Net Price(Total Value)Index,and uses the ADCC-MIDAS model for empirical research.The sample interval is from January4,2007 to December 28,2018.After using ADCC-MIDAS model to describe the dynamic correlation between stocks and bonds in China,we find that the correlation between stocks and bonds has a very significant time-varying characteristic,and the short-term components fluctuations of stocks and debts have synergistic changing trend with long-term components.And the correlationbetween stock market and bond market has an asymmetrical effect that the negative impact effect is greater than the positive impact effect.Secondly,after obtaining the long-term components of correlation,based on flight to quality and macro-factor driving,using the VAR model to analyze the macro factors and market factors affecting the correlation of the stock-bond market.Through the VAR model impulse response analysis and variance decomposition,we find that: The long-term correlation of stock and bond is mainly affected by the macroeconomic factors.The contribution of industrial added value,interest rates,housing price and money supply is higher,but the contribution of inflation rate,stock market volatility,stock market liquidity and economic policy uncertainty is smaller.Macro factors are the main factors affecting the long-term correlation of stock-bond.The indicators related to asset security transfer have insufficient ability to explain market correlation changes.This may be related to the retailers’ large proportion in stock market and the investor structure of the inter-bank bond market dominated by commercial banks.The investor structure may cause funds cannot flow freely between the stock market and the bond market,resulting in low ability of asset safety transfer to explain market correlation.Finally,this study proposes relevant policy recommendations.Investors should pay attention to the trend of correlation and the asymmetry of the positive and negative impact of new information,which helps to adjust the asset allocation ratio and risk position.The regulatory authorities should improve the bond market system and clear the channels for capital circulation to play the risk-avoidance function of the bond market better.When formulating and implementing macroeconomic policies,the regulatory authorities should consider the impact on the correlation of the stock-bond market and maintain the stability of the policy.When macroeconomic factors fluctuate greatly,measures should be taken in time to strengthen the regulation of the bond market to promote the coordinated and healthy development of the two markets.The research innovations in this paper are mainly reflected in the following aspects: First,a new ADCC-MIDAS model is proposed,which can extract the long-term and short-term components of correlation,and reflect the asymmetric effect of positive and negative shocks.Secondly,use the long-term components of the ADCC-MIDAS model extraction,through the VAR model,to further analyzethe macro and market factors of affecting the stock-bond market correlation.