The Research On The Relationship Of Volatility Shift And Dynamic Conditional Correlation Of Bond,Stock And Foreign Exchange Markets

The research based on the idea of the research of dynamic correlation analysis of financial crisis study on the relationship between volatility clustering and dynamic correlation.The major target of this paper is the bond,stock and foreign currency markets,and research on the relationship between volatility clustering of bond,stock and foreign currency markets and corresponding dynamic correlation.This paper use VAR-FIAPARCH-cDCC model as the main model.First of all,the VAR-FIAPARCH-cDCC model mean equation take VAR model,which can filter the auo-correlation and interaction across lag correlation.Secondly,the variance equation take FIAPARCH model,whichis very flexible and can capture several volatility properties such as long memory,asymmetry,leverage effect and kurtosis.And then,we use cDCC model proposed by Aielli(2013)to estimate the dynamic conditional correlation between markets,instead of DCC proposed by Engle.Based on those,we use the penalized contrast function proposed by MarL(2003)to estimate the volatility clustering points,and analysis the relationship with dynamic conditional correlation,and then,to testify the relationship by constructing regression model,and give the economical explanation.Lastly,we study the America financial markets using the same research methods,and compare the results,and analysis its.The empirical study show that there are four kinds of phenomenon:First of all,the dynamic conditional correlation will abruptly change at the point of volatility shift;secondly,the dynamic conditional correlation have positive and negative change at the point of volatility;thirdly,the abrupt change of dynamic conditional correlation at the time of volatility shift will not persist too long,otherwise it's temporary;At last,the dynamic conditional correlation have different degree reaction to different volatility shift.We comfier its existence by testifying the regression model.Compare to Chinese financial markets,we find that those phenomenon behave significantly in America financial markets...