The Study Of Foreign Exchange Market Volatility And Risk Calculation Analyse

As economic globalizing, more and more trades happen among countries, so the risk of exchange rate price volatility becomes the most important topic among these traders. Risk arose from the volatility of asset price or value, if to go in depth study risk, the first thing to do is to study market's volatility. In the first part of last century, exchange rate pricing theoretics was countries exchange policy basic theoretics, so these there were some helpful and guidance for the exchange rate study and practice. As the study goes in depth, there is much restrict in the hypothesis, so the exchange rate pricing theoretics is not practical. The sixties of last century, the effect market hypothesis (EMH) was created, was popular in financial market study in that years, many studies apply the EMH in real market practicing. After EMH was created, many scholars use many models to prove EMH result is correct, but the prove was not satisfied, the hypothesis are too strict, is not practical in real market. In the end, EMH is not of persuasion in volatility theoretics study or market practice.In 1982, Engle created one variability model—ARCH model, in this 20years, the ARCM model was developed into a big model family by many scholars. These models can describe the fat-tail, high-peak and volatility clusting of financial time series visualize, it's also applied widely in practice. This pepper apply ARCH model to compute three exchange rate price time series, GBP/USD,USD/CNY and USD/CNY time series. There are fat-tail, high-peak, and ARCH effect, but there is not lever effect. In GBP/USD,USD/CNY time series, there is strong long duration information impact, but EUR/USD time series has strong shorten information impact. And it's also found that abnormity in the USD/CNY time series, it maybe a highlight of this paper. In the last part of pepper, the conditional variance is import to calculate VaR, then examine and analyse, VaR can makes risk management into figure management, so the risk surveillance becomes maneuverability risk management. One of the key of model calculation is to choose the distribution of variance, the better distribution is T- distribution. The second key of model calculation is choose the VaR confidence lever, and there is not a standard lever in theoretics study or practice, there is further study to go. To use the VaR calculate risk, we can find worse result in USD/CNY time series, in the EUR/USD, GBP/USD time series there are better result.