Copula-CoVaR Systemic Risk Assessment Based On Multiple Methods

As the close connection between financial markets continues to strengthen,accurate measurement of systemic risks has become an important basis for relevant financial practitioners and decision makers to make reasonable choices and decisions.It is very important to establish a scientific and effective mathematical model to measure the systemic risk of financial markets.Based on the existing risk measurement and related research,the VaR method can only independently measure the market’s risk status,resulting in underestimation of risk,and cannot effectively estimate systemic risks between financial markets.This article uses the CoVaR method to evaluate systemic risks in different stock markets.This method has better universality,covering various factors that affect the financial market,while taking into account the non-linearities in the time series of each financial market.Due to the complexity,diversification and time-varying characteristics of financial markets,the interdependence between financial markets in different periods is also constantly changing,especially when the financial market is in a downturn,the correlation between different financial markets will be significantly enhanced.In this paper,Copula function is used to describe the correlation.Copula function is an effective modeling method,which can not only reflect the linear and nonlinear and symmetric and asymmetric correlations in different financial markets,but also capture the tail dependence between financial markets.This paper constructs two types of models: dynamic GARCH-Copula-CoVaR model and dynamic SETAR-Vine-Copula-CoVaR model to study systemic risk in different situations.Specifically,the dynamic GARCH-Copula-CoVaR model first uses the GARCH model to fit the marginal distribution of the time series of returns,and then uses the Copula function to characterize the correlation and joint distribution between the two financial markets.Finally,the premise of the first two steps The solution of CoVaR is carried out below,and it will be extended to dynamic situations in the subsequent work.Second,the dynamic SETAR-Vine-Copula-CoVaR model divides the state of the financial market in the case of the first model and expands the research object to high-dimensional situations.Specifically,first use SETAR to divide the state of the financial market Then,the divided states are combined to establish a CoVaR solution for each combination.In the solution process,Vine-Copula is used to reduce the dimensionality of high-dimensional variables and the high-dimensional CoVaR solution is used.In the follow-up work,it is also extended to dynamic situations.For model one,select the Shanghai Stock Index and the six sub-financial institutions in the Shanghai Stock Index,analyze the systemic risks of the sub-financial market on the Shanghai Stock Index,and find that all six sub-financial institutions have a significant positive financial risk effect on the Shanghai Stock Index.But the performance is different,and the cluster analysis of the six selected sub-financial institutions still has the same conclusion.For model two,the Shanghai Stock Index and the three sub-financial institutions with the highest market capitalization and different operating characteristics in the Shanghai Stock Index were selected as the research objects,and different conclusions were drawn under different financial market conditions.Finally,this article also gives relevant suggestions to provide a strong reference for financial regulators.