Foreign Exchange Market Volatility And Var Risk Measure Research

With the collapse of the price system of fixed exchange rate and the rise ofworldwide financial liberalisation, the financial institutions and foreign enterprises, as themain body of the foreign exchange market, are facing unprecedented increasing exchangerate risks. The econometric model of the family of GARCH model and the VaR calculationmethod is applied to the analysis of exchange rate volatility characteristics and thecalculation of the risk of exchange rate fluctuations, it will help us to reveal the RMBexchange rate volatility characteristics of the exchange market, at the same time it willalso provide a theoretical and empirical results support for the major commercial banksand regulatory decision-making.Firstly, this article describes the mainstream of exchange rate volatility VaR riskmeasurement methods, taking into account the estimated VaR is closely linked with theextreme value distribution, I explore the extremes of VaR calculation.Secondly, in terms of the distribution of exchange rate returns, the statistical resultsshow that the yield rate sequences all have a fat tail and non-normality characteristics.From volatility modeling point of view, I then examine that the exchange rate volatilitysequence has fluctuation persistent, heteroscedasticity and leverage effect characteristics.As to the aspects of the empirical analysis, this paper discusses the applicable premise forGARCH process, then I give the maximum likelihood estimation results of the GARCHmodel, while the error term is based on the normal distribution, t distribution andgeneralized error distribution. Considering the usual information criteria do not reflect themerits of the data fitting GARCH process, I put forward the improvement of originalinformation criterion, and pick out the best fitting model using the modified informationcriterion.Finally, in order to effectively describe our exposure to fluctuations in the foreignexchange market rates, I use the different calculation methods of VaR to measure thevolatility risk, I then use the back-testing methods to test the fit of the actual yieldsequence to the sequence based on the estimated results of different methods of calculating VaR. On this basis, through the study of GARCH-M model, I explore therelationship between the foreign exchange investment return and time-varying risk, I findthe best measurement for time-varying risk of different yield sequences.