Empirical Study On The Different Stages Of The Commercial Bank Exchange Rate Risk Measure

The national currency the IMF on November 30,2015 announced the inclusion of the RMB IMF SDR currency basket, the decision will become effective October 1,2016.Thus, the process of internationalization of the RMB go further, in the current complicated international financial markets, currency exchange rate fluctuations have caused more than mere appreciation of the RMB to depreciate, also related to the value of the entire national wealth, competitiveness in international trade and the national security of the financial system. China’s commercial banks to exchange rate risk management started late, the exchange rate risk measurement is relatively backward, according to the 2014 annual report of the major commercial banks, China’s major commercial banks mainly used the 99% confidence interval and historical simulation method to calculate the holding period of 1 day value at risk. Changing the degree of China’s market economy deepened, exchange rate risk exposure is gradually becoming diverse, commercial banks face a variety of exchange rate risk, but also the use of the portfolio to earn money, VaR method may better overall measure of market risk.This paper mainly take VaR method based on the RMB exchange rate fluctuations at different stages of the commercial banks to exchange rate risk empirical research GARCH Models and Monte Carlo Simulation. In this paper, the different time segments with the former, after the exchange rate reform in 2005 to 29 May 2015 the dollar and euro against the 10 years between the yuan central parity to study depending on the economic situation and exchange rate policy is divided into three stages, Further comparison of different economic backgrounds choose different methods to accurately measure the exchange rate risk. The first phase began in 2005 after the exchange rate reform in July 25,2008 to June 30; the second phase is July 2008 1-July 19,2010, during the global financial crisis continues to worsen, my Government to stabilize the economy, in two years time will be in control against the dollar around 6.83 yuan per dollar; the third stage July 20,2010 to May 29,2015, the Chinese government in order to advance RMB internationalization strategy, July 19,2010, signed the revised "Hong Kong banks RMB business clearing agreement", the RMB internationalization process accelerated.Through three different stages of the empirical analysis, the following conclusions:the first phase of dollar a better fit model is GARCH(1,1)-M model; the second stage better model is under the generalized error distribution.the EGARCH (1,1)-M model, asymmetric effect coefficient EGARCH-M model is significantly different from 0, indicating the presence of leverage at a critical time national control of economic crisis exchange rate of return sequence and significance tests show yields tail sequence than normal but thicker than the thin t distribution; third stage is a better fit GED under GARCH (1,1)-M model, t distribution under GARCH (1,1)-M model does not pass the significance test, described GED can better fit the series.2. Model euro third stage of the final choice is EGACH generalized error distribution model under asymmetric effect coefficient model is significantly non-zero, indicating the presence of the leverage effect, the model coefficients t Distribution was not significant, indicating that the third phase of the euro return series features more consistent with the GED. And because the first and second phase of the euro and the RMB can not deal directly, so the model fits worse results.3. Monte Carlo simulation results from different confidence level found that as confidence increases, the risk was gradually increasing trend; calculated VaR value of the third stage according to two methods Kupiec test results indicate that, in not possible when the path of a large number of Monte Carlo simulations, GARCH family model and Monte Carlo simulation method accuracy was essentially flat. Finally, this paper empirical results make relevant recommendations based on the analysis.