The Study On The Construction Of Exchange Rate Volatility Models And The Measurement Of Value At Risk

Grasping the characteristics of exchange rate fluctuations accurately is essential for foreign investment, foreign exchange risk management and prevention. It has a huge effect on the national macroeconomic and microeconomic. Since July 21,2005, China has begun to implement a market-based, managed floating exchange rate regime with reference to a basket of currencies. The multilateral exchange rates of RMB will be calculated based on market supply and demand and with reference to a basket of currencies. The frequency and amplitude of exchange rate fluctuations will greatly be enhanced. With a correct understanding of the regular pattern of RMB exchange rate fluctuations, we can not only improve the level of exchange rate management, and can reduce the risk of exchange rate fluctuations to the national economy and the individual economies in contrast with international exchange rate. It has great theoretical and practical significance to sustained, stable and healthy development of the national economy.In this paper, the objects of study are respectively Canadian dollar, Japanese yen, British pound against U.S. dollar exchange rate return series and RMB against U.S. dollar, euro, Japanese yen exchange rate return series after exchange rate reform in 2005, with emphasis on depth study of exchange rate volatility models'construction and VaR's measurement. Exchange rate volatility models include the GARCH type models (GARCH, IGARCH, EGARCH) and fractal difference models (FIGARCH, FIEGARCH), and are assessed from two aspects, the goodness of fit and the effect of prediction. Modeling results show that the effect of models'quality should be analyzed combining with the practical issues. The test results based on VaR measurement also show that good or bad effect of forecast depends on specific conditions. Different time series, different levels of confidence, results are also different. The fractal difference models are not necessarily optimal. The main reasons for these results are the different internal mechanism of models and the different selected samples.After exchange rate reform, society is particularly concerned about the marketing process of exchange rate. This article analyzes the effectiveness of reform from two respects, trend of exchange rate fluctuations and model salaction.wo can get that exchange rate is deeply marketed after exchange rate reform.