Research On The Financial Crisis Contagion Between China And The U.S. Based On Copula And VAR Model

The rapid development of modern science and technology as well as global economic integration process have accelerated the whole global economy.Both the real economy and the financial sector have reached the peak.Although all the countries and regions enjoy the dividends brought by economic integration,they can't avoid systemic risks.For example,when a country or region experiences a financial crisis,all the other countries or regions will be affected in different levels and serious economic crisis may even emerge.At the same time,with the political connections,trade exchanges and economic intercommunication between countries more complicated,it's impossible to explain economic relations between countries or regions with traditional economic theories.What's more,the multi-path and extensiveness feature of financial crisis contagion makes it difficult to study financial crisis contagious.At present,there are many problems in academic researches on financial crisis contagion.For example,the non-dynamic nature of empirical methods often results in large differences between research results and actual situations.This paper demonstrates the effects of financial crisis contagion between China and the United States from two perspectives.One is the dependency relationship based on the Copula model and the other is the causality based on the VAR model.Based on the Copula model to verify the contagion effect of the financial crisis in China and the United States: This paper selects the representative stock market indices S&P500 and CSI300 in the United States and China: using the fractal market method to estimate the volatility of each market separately,and standard income after eliminating the heteroscedasticity.The rate series is transformed into a uniform distribution over the interval by the probabilistic integral transform.Second,in order to clearly show the dependency relationship between the US stock market and the Chinese stock market,especially the tail dependency,three kinds of static Copula models are used to model the dependency relationship.A comparative analysis of empirical results was conducted to find the model SJC-Copula with the best fitting effect and to verify the existence of significant interdependence between China and the United States.Next,a time-varying SJC-Copula model was constructed to analyze the changes in the "non-hazardous period" and the "dangerous period" tail dependency of the financial crisis,demonstrating the contagion effect of the financial crisis between China and the United States.In order to examine the contagion effect of the financial crisis between Chinaand the United States from the perspective of causality,the VAR model continues to be constructed: S&P500 and CSI300 are selected as samples,and time series are still divided into "danger period" and "non-danger period" of financial crisis.Building a VAR model on the basis of smooth time series and analyzing the Granger causality test results: In the “non-hazardous period” of financial crisis,the US stock index volatility is the Granger cause of China's stock index volatility;during the“dangerous period” of the financial crisis,The changes in the stock indices of the two countries are Granger's reasons,and the contagion effects of the financial crisis in China and the United States have been found.Next,an impulse response analysis of the time series was performed to compare the changes in the “danger period” and“non-danger period” impulse responses of the financial crisis.It was found that the“danger period” of the financial crisis was significantly higher than that of the“non-danger period”.The contagion effect of the financial crisis between China and the United States was found.