Combining With The R/S Analysis Of A Variety Of VaR Risk Measurement Of RMB Exchange Rate

The exchange rate is the core economic variable, It is not only adjusting the macro andmicro economy but also affecting the equilibrium of home and abroad. With the newly reformof the RMB exchange rate on21, July2005, exchange rate fluctuation is gradually marketing.We are going to make an empirical risk analysis about RMB exchange rate, The articlefocuses on the following aspects:This paper first reviews the development course of RMB exchange rate and riskmanagement status, theoretically comparing several traditional financial risk measurementmethods, so providing background support for the following better risk measure.Secondly, in order to improve the standard of exchange rate risk measurement based onVaR, We will make the random test, normal test and heteroscedasticity test of the geometricreturn rate of RMB exchange rate. and then uses the R/S fractal method to analysisthe long-term memory period of RMB exchange rate (the USA/RMB, EUR/RMB, HKD/RMB, yen/RMB, GBP/RMB), to determine the long-memory cycle and mean cycle, so Wewill have two kinds of study period in the VaR risk measure.Then, the key part of this article, We plan to use three models including historicalsimulation, variance-covariance method and GARCH-VaR model to measure VaRrespectively, We also let it in different confidence levels and study period, according to theaccuracy test shows that: GARCH-VaR model is the optimal measure effect, and mostimportantly, The RMB exchange rate risk measure more accurately when combined withfractal analysis of long-memory cycle periodFinally, this paper are summarized based on the empirical result, and from three partsuch as tool selection, method of application to put forward policy suggestions about howto build VaR as the risk measure standard of the exchange rate risk measurement system andput forward policy suggestions.