Analysis Exchange Rates Dependence Based On Dynamic Copula-ARMA-GJR Model

Studying the co-movements across financial markets is an important issue for riskmanagement and portfolio management. There is a great deal of research focusing onthe co-movements of international equity markets rather than the foreign exchangemarkets. But the foreign exchange market is an important segment for the wholefinancial market, so studying the dependence between the exchange rates is of greatsignificance. Especially since July21,2005, RMB is indeed in the process of openingup after the exchange rate system reform, which results in the increase of uncertainty ofRMB exchange rate fluctuations and foreign exchange risk. Meanwhile, recently thefinancial crisis frequently took place, and the exchange rates of Europe and Americaexperienced sharp fluctuation. The U.S.-led nations frequently put pressure on theappreciation of the RMB to relieve the pressure of their domestic economic recoveryand growth. In such a situation, we should strengthen foreign exchange riskmanagement urgently, and it is significant important for foreign exchange riskmanagement to accurately depict the dependence between the exchange rates. Therefore,it becomes more urgent to study the dependence between RMB exchange rates.It is difficult to conduct the studies on the dependence between the RMB exchangerates, and this is mainly because the central bank controls the middle price of the foreignexchange, then the foreign exchange market is not truly free owing to the influence ofthe noise. So it is a great challenge for models choosing in RMB exchange rates studies,and that the single dynamic Copula model is unable to accurately depict the dependencebetween the RMB exchange rates. In this paper, we developed multiple dynamicCopula-ARMA-GJR models to conduct an empirical study on the dependence structurebetween RMB exchange rates according to the high-peak-fat-tail and asymmetricfeatures of financial return assets exhibit.The empirical results show that: there exists significant negative dependencebetween the exchange rates of USD/RMB and EUR/RMB, USD/RMB and USD/RMB,USD/RMB and USD/GBP, and that the negative dependence between the exchangerates of USD/RMB and EUR/RMB which is the most significant. But comparing withabove situations，there exists significant positive dependence between the exchangerates of EUR/RMB and GBP/RMB, meanwhile, the dependence between the exchangerates of JPY/RMB and EUR/RMB, JPY/RMB and GBP/RMB are sometimes positive and sometimes negative. What’s more, the dependence underwent a great change whileit is different from the positive dependence in stock markets which enhanced in thefinancial crisis or other extreme events. Additionally，when it comes to tail dependence,the upper and lower dependence between the exchange rates of USD/RMB andEUR/RMB, USD/RMB and JPY/RMB, USD/RMB and GBP/RMB both are closed tozero, which implies there not exist the probability of soaring or plummeting in the sametime, while the upper and lower dependence between the exchange rates of EUR/RMBand JPY/RMB appears a strong volatility, which suggests there exists risk contagionrelationship between them.