Dynamic Analysis Of China’s Foreign Exchange Market Based On Markov Regime Switching Model

During the30years, China’s exchange rate arrangement has been adjusted and reformed many times. The RMB exchange rate also has experienced many complicated changes. In1994,the RMB unofficially pegged to the U.S. dollar, but only in the U.S. dollar against the RMB exchange rate fluctuated between8.27to the8.28yuan. In2005, the RMB exchange rate based on market supply and demand, reference to a basket of currencies, the managed floating exchange rate system. In2008, the United States subprime mortgage crisis triggered by the global financial crisis, China adjust the RMB fluctuations in response to the crisis. In2010, Bank of China said it will further advance the reform of the RMB exchange rate formation mechanism. For these reasons for the fluctuation of the RMB exchange rate uncertainty, therefore, the RMB exchange rate behavior becomes particularly difficult. In recent years, the RMB appreciation pressures continue to increase, the changes of China’s foreign exchange market has become more and more important.This paper based on this background, the Markov switching model and the GARCH model as the theoretical basis, the maximum likelihood estimation method using. At first, using the GARCH model, EGARCH model and GJR model to estimate the Chinese foreign exchange market volatility, confirmed the non symmetry of its fluctuation, which is bad news and good news have different impact effect. Then set up MRS-GARCH model volatility on China’s foreign exchange market for further analysis, the Chinese foreign exchange market volatility is divided into high volatility and low volatility state two state, the fluctuation of the two states were fitted by the Markoff state transition process, and get the transition probability and state probability. During the analysis process, according to the sequence shown partial peak thick tail distribution, the model assumes that the residuals distribution are different:the normal distribution, t distribution, time-varying t distribution and generalized error distribution. Finally, according to the AIC, BIC standards and the contrast loss function, carries on the comparison to each model under different distribution, selects the relatively optimal model was further analysis.This paper is based on the research of the RMB exchange rate returns, and introducing another sequence, namely the RMB non deliverable forward contracts offer returns as comparative data, compared to market behavior to model, finally get the conclusions:1, China’s foreign exchange market volatility has obvious nonlinear characteristic; Ⅱ. China’s foreign exchange market volatility asymmetry, the size of the shock is good news and bad news on the foreign exchange market is not the same; two state significant transfer process exists the Chinese foreign exchange market, in the two state transition probability is higher, while in the steady state probability and low volatility was significantly lower than in the steady state probability under the condition of high volatility. Although NDF RMB yield considering the impact of market behavior, but from the MRS_GARCH model analysis of optimal distribution found in influencing market behavior on the model result is not very obvious, parameters of two sequences with similarity.According to the empirical conclusions, combined with the actual characteristics of China’s foreign exchange market, in the end the three angle to the government, financial institutions and enterprises, a simple analysis and suggestions on the reform of China’s foreign exchange market, legal system, derivatives and risk aversion and other aspects, so as to establish a more mature the stable foreign exchange market.