The Research On RMB Exchange Rate Risk Measurement Based On Bilinear GARCH-CVaR Model

With the continuous improvement of China's status in the world,the trade between China with other countries and regions in the world has gradually increased,and the RMB also plays an increasingly important role in the international economic arena.As an important economic adjustment variable and policy adjustment tool,the RMB exchange rate is also an important hub for coordinating balanced domestic and international economic development.The fluctuation of exchange rate data brings risks to the corresponding investment activities,and accurate measurement of exchange rate investment risk will help investment entities to make correct decisions.In order to adapt to economic development in different periods,the exchange rate system needs to be adjusted accordingly.Different exchange rate systems will cause different fluctuations in exchange rate data.The new RMB exchange rate system has increased the fluctuation range of exchange rate data.Exchange rate changes contain more uncertain factors and investment risks have further increased,and the accurate measurement of exchange rate risks is even more important.Based on the analysis of financial time series and exchange rate risk-related theories,this article uses the ARCH family model,value-at-risk(VaR),and conditional value-at-risk(CVaR)models to fit and analyze exchange rate data,and comprehensively compares the basis of the two methods.Based on the above,the conditional value-at-risk method can more accurately measure the risk of investment,and according to the results,put forward some countermeasures and suggestions to investment entities from different aspects.This paper mainly includes the following two aspects:First,the ARCH family model is established under different distributions by using the historical data of RMB to US dollar exchange rate.After comprehensive comparative analysis of several models,the parameter estimates of the model are given.According to the AIC criterion,the bilinear GARCH model has the best fitting effect on the data under the t-distribution,and the subsequent calculation of risk value is carried out on this basis.Second,Combined with the parameter estimation results of bilinear GARCH model under t distribution,VaR and CVaR models are used to measure the exchange rate risk of RMB at the same time.After the "new exchange rate reform",the data of RMB exchange rate fluctuates more widely and it is more difficult to fit.The results of empirical comparative analysis show that CVaR model can measure the risk of RMB exchange rate more accurately.Finally,based on the results of empirical research,it puts forward relevant opinions and suggestions from the perspective of market,tools and institutions.