本文旨在研究金融商品的市場風險，以風險值（VaR值）衡量市場風險，並通過回溯測試檢驗衡量因市場風險引發極端損失的模型與方法是否有效。首先由定義金融商品曝險（股指期貨、遠期外匯和歐式選擇權）的損失分配公式出發，分析商品的市場風險因數並設立對應的模型。本文以 EGARCH(1,1)模型對股價指數風險因數建模；以Dynamic Nelson-Siegal 利率期限結構模型（DNS 模型）對利率風險因數建模；以隨機遊走模型(random walk)模型對匯率風險因數建模。接著使用蒙地卡羅模擬法對風險因數進行模擬，再進一步推導出金融商品的風險值。最後以回溯測試，kupiec test檢驗本文所用的VaR值方法是否有效。研究中選擇以滬深300股指期貨、人民幣兌美元遠期外匯與S&P500股指歐式選擇權為實證數據的研究對象。實證結果顯示，本文使用的計算VaR值的方法通過模型檢驗，均能有效衡量各金融商品的市場風險，為建立以資本計提為核心的市場風險管理系統打下堅實的基礎。

This paper put emphasis on market risk of financial products, by using the method of Value at Risk (VaR). Through backtesting technique to prove that the models and methods used in this paper for measuring extreme loss caused by market risk for financial products are effective. Firstly, loss distribution formula of risk exposure is defined for stock index future, foreign exchange forward and European options. Different risk factors are identified, and appropriate models are selected for risk factors. EGARCH(1,1) model is used for building risk factors of stock indexes; Dynamic Nelson-Siegal (DNS model) is selected for fitting interest rate risk factors; random walk model is used for building risk factors of foreign exchange rates. Then, Monte Carlo simulation is used to simulate all risk factors, and VaR of financial products can be derived. At last, kupiec test is used to test whether the models and methods for calculating VaR are suitable or not. Shanghai and Shenzhen 300 Stock Index future, RMB to USD foreign exchange forward and S&P500 Stock Index European options are selected for empirical research. The results show that models and methods used for VaR all passed the kupiec test, which means that market risk can be measured properly. VaR approaches defined in this paper have built solid foundations for a market risk management system taking capital charges as the core.

Black, F., Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of political economy, 81(3), 637-654.

Vasicek, O. (1977). An equilibrium characterization of the term structure. Journal of financial economics, 5(2), 177-188.

Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Economatrica, 50(4), 987-1008

Meese, R. A., Rogoff, K. (1983). Empirical exchange rate models of the seventies. Journal of international economics, 14(1), 3-24.

Cox, J. C. , Ingersoll, J.E., Ross, S.A. (1985). A theory of the term structure of interest rates. Econometrica (1986-1998), 53(2), 385-407.

Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of econometrics, 307-327.

Nelson, C. R., Siegel, A.F. (1987). Parsimonious modeling of yield curves. The journal of business, 60(4), 473-489.

Nelson, D. B. (1991). Conditional heteroskedasticity in asset returns: a new approach. Econometrica (1986-1998), 59(2), 347-370.

Diebold, F. X., Li, C. (2006). Forecasting the term structure of government bond yields. Journal of econometrics, 130(2), 337-364.

封毅，盛雷，楊青山.(2007). 基於EGARCH模型的中國股市波動性的實證研究. 金融經濟，2007（24），124-125

Nieppola, O. (2009). Backtesting Value-at-Risk models. Master’s thesis in Economics, Helsinki School of Economics
康書隆，王志強. (2010). 中國國債利率期限結構的風險特徵及其內涵資訊研究. 世界經濟，2010（7），121-243

陳平，李凱.(2010). 人民幣匯率與宏觀基本面：來自匯改後的證據. 世界經濟，2010（9），28-45

Hull, J. (2012). Risk management and financial institutions. Hoboken: John Wiley & Sons, 183-186

吳佳興.（2013）銀行利率風險與金融商品市場風險之實證分析與應用

Engel, C., Mark, N. C., West, K. D. (2015). Factor model forecasts of exchange rates. Econometric reviews, 34(1-2), 32-55.