如何妥善的管理風險是非常重要的，我們根據資料為2002年1月1日至2004年12月31日七大工業國(G7)之主要貨幣兌換美元的即期匯率，包括日圓、英鎊、法國法郎、義大利里拉、加拿大幣及歐元為實證對象，針對即期匯率波動的風險值作探討，比較變異數-共變異數法、歷史模擬法、蒙地卡羅模擬法、混合常態分配與極值理論。本研究主要目的在比較何種方法在估計匯率波動風險值時較具準確性，希望能夠提供一客觀風險揭露的參考工具，讓必須面臨匯率風險的企業及金融機構能有效判斷目前的風險程度，以採取必要之避險措施。以往的文獻，大多討論傳統模型，其餘的部份一些針對混合分配做探討，或是只針對極端值，為改善以往的文獻，本研究將所有方法統合在一起，其比較結果應更具全面性。

實證結果發現，使用歷史模擬法皆得到不太好的表現，因為歷史資料可能代表性不夠，由於歷史模擬法只有在資料結構無明顯變化時，才有較佳的預測力；其損失超過風險值之頻率最不接近α，這或許是波動性具群聚的效果，使得沒有考慮到近期資料對未來報酬影響的效果較顯著的歷史模擬法表現較差。以Hull and White(1998)方法所估計之混合常態，可以掌握到資料的特性。加拿大幣兌美元匯率屬於厚尾分配，所以使用極值分配來估計風險較為準確。

Financial disasters have emphasized the importance of effective risk management for financial institutions. We collected the data of U.S. Dollar against Japanese Yen, British Pound, French Franc, Italian Lira, Canadian Dollar, and Euro Dollar as our empirical data from January 1, 2002 to December 31,2004 to analysis and compare the accuracy of estimating Value at Risk (VaR) under the models of variance-covariance method, historical simulation method, monte carlo simulation method. Since financial asset returns don’t often follow the hypothesis of normal distribution, we try to evaluate VaR by mixture of normal distribution and extreme value theory. The purpose of my study uses various Value at Risk (VaR) models for monitoring the market risk of foreign investment port folio. According to the different kind of models, we try to find an objective risk-disclosing way by comparing the performances of these methods.

The empirical results are as follows: In general, mixture of normal distribution has a good ability to forecast risk. Historical simulation method is not a good way to mesure risk if the structure of data change.

中文部分：
1.李進生、謝文良、林允永、盧陽正，民國2001，「風險管理(VaR)理論與運用」，清蔚科技出版社。
英文部分：
1. Alexander, C.O. and Leigh, C.T.,“On the Covariance Matrices Used in Value at Risk Models,”Journal of Derivatives, 1997, Vol.4, pages 50-62.
2.Beder, T.,“VAR：Seductive But Dangerous,”Financial Analysis Journal, 1995, Vol.51, pages 12-24.
3.Bollerslev, T.,“Generalized Autoregressive Conditional Heteroskedasticity,”Journal of Econometrics, 1986, Vol.31, pages 307-327.
4.Bradley B.O. and Taqqu M.S.,“Financial Risk and Heavy Tails,”To appear in the volume Heavy-tailed distribution in Finance, Svetlozar T.Rachev, editor, North Holland, 2001.
5.Danielsson, Jon, and Casper de Vries,“Value-at-Risk and Extreme Returns,”LSE Financial Markets Group Discussion Paper 273,London School of Economics, 1997.
6.Duffie, D. and Pan, J.,“An Overview of Value at Risk,”Journal of Derivatives, 1997, Vol.4, pages 7-49.
7.Eagle, R.,“Autoregressive Conditional Heteroskedasticity with Estimates of the Variance of United Kingdom Inflation,”Econometrica, 1982, Vol.50, pages 987-1007.
8.El-Jahel, L., Perraudin, W. and Sellin, P.,“Value at Risk for Derivatives,”Journal of Derivatives, 1999, Vol.6, pages 7-26.
9.Embrechts, P.,Claudia Klupperlberg, and Thomas Mikosch,“Modelling Extremal Events for Insurance and Finance,” Springer, Berlin, Germany, 1997.
10.Golub, Bennett W. and Leo M. Tilman.,“Measuring Yield Curve Risk Using Principal Components Analysis, Value at Risk, and Key Rate Durations,”Journal of Portfolio Management, 1997, Vol.23, pages 72-84.
11.Hall, P.,“Using the Bootstrap to Estimate Mean Squared Error and Select Smoothing Nonparametric Problems,”Journal of multivariate Analysis, 1990, Vol.32, pages 177-203.
12.Hamilton, J.,“A Quasi-Bayesian Approach to Estimating Parameters for Misxture of Normal Distributions,”Journal of Business and Economics Statistics, 1991, Vol.9, pages 27-39.
13.Hendricks, D.,“Evaluation of Value-at-Risk Models Using Historical Data,”Economic Policy Review, 1996, pages 39-70.
14.Hill, B.M.,“A Simple General Approach to Inference about the Tail of Distribution,”Annuals of Statistics, 1975.
15.Ho, T.,“Key Rate Durations：Measures of Interest Rate Risks,” Journal of Fixed Income, 1992, Vol.2, pages 29-44.
16.Hull, John and Alan White,“Value at Risk when Daily Changes in Market Variables Are not Normally Distributed,” Journal of Derivatives, 1998, Vol.5, pages 9-19.
17.Jorion, Philippe,“Value at Risk：The New Benchmark for Managing Financial Risk,” 2001.
18.J.P.Morgan,“Riskmetrics Technical Manual,”1995.
19.Longin, Francois,“From Value at Risk to Stress Testing：The Extreme Value Approach,”Journal of Banking and Finance, 2000, Vol.24, pages 1097-1130.
20.Pickands, J.,“Statistical Inference Using Extreme Order Statistics,”The Annuals of Statistics, 1975, Vol.3, pages 119-131.
21.Thomas S.Y.Ho, Micheal Z.H.Chen, and Fred H.T.Eng,“VaR Analytics：Portfolio Structure, Key Rate Convexities, and VaR Betas,”Journal of Portfolio Management, 1996.
22.Venkataraman, S.,“Value at Risk for A Mixture of Normal Distributions：The Use of Quasi-Bayesian Estimation Techniques,” Economic Perspectives, 1997, Vol.21, pages 2-13.