在現在的學術上，有許多研究方法來預測金融危機。而許多金融機構也採用了代表性的指標來預測危機。這些方法和指標雖然不可能直接評估，但還是以不同的估計方式估來呈現，並朝各方面的發展。儘管，至今仍無法證明哪一方法或指標最為適當，我們仍然試圖找到具有某一特性的產業能幫助我們在股票危機發生前提前預警做出適當地應變措施。
在本文中，我們採用的資料為S＆P100，時間期間為1977年1月至2008年12的月資料。我們先將資料透過Fama-French三因子加上動能因子和投資人情緒做資料上的處理，接著透過集群分析分成四個組群。最後，我們採用GARCH-SGT模型，並運用VAR來預測股市危機。
本文中，我們發現預測股市危機的關鍵因素是峰態值，越高的峰態值越具有預測的能力而非高的波動性。此外，當產業具有較大規模這項特性時，也較有可以預測股市危機的能力。另一方面，我們也可以透過此模型來做到確認的目的，藉此，我們可以在危機發前做適當地風險控管以降低損失。

There are several methods to predict financial crises. There are also several types of indicators used by financial institutions. These indicators, which are estimated in different ways, often show various developments, although it is not possible to directly assess which is the most suitable. Here, we still try to find what characteristics that industry group has and forecast financial crises
In this paper, our data started from monthly of 1977 January to 2008 December in S&P100. We consider Fama-French and Cluster Analysis to process data to make data with same characteristic within a group. Then, we use GARCH type models and apply it to VaR predicting stock turmoil.
In conclusion, we found that the group which has high kurtosis value is the key factor for predicting stock crises instead of volatility. Moreover, the characteristics of this industry which can predict stock crises is a great scale. On the other hand, we can through this model to double check the reaction for anticipating. Therefore, people can do some actions to control risk to reduce the loss.

論文審定書 i
摘要 ii
Abstract iii
Tables v
Figures vi
1. Introduction 1
2. Method 6
2.1 GARCH (1, 1) model with skewed generalized t distribution (GARCH-SGT) 6
2.2 skewed general t density 6
2.3 Measurement and evaluation for distribution-based VaR models 8
2.3.1 Definition and estimation 8
2.3.2 Conditional-SGT-VaR approach 8
2.4 Evaluating VaR performance 8
2.4.1 Unconditional coverage test (LRuc) 9
2.4.2 Conditional coverage test (LRcc) 9
3. Data 11
4. Empirical result 12
4.1 Result of stock market 12
4.2 Evaluting the performance of VaR 17
4.3 Analysis of clustering 17
5. Conclusion 19
Reference 21

A.Cipollini and G. Kapetanios, 2009, Forecasting financial crises and contagion in Asia using dynamic factor analysis, Journal of Empirical Finance 16, 188–200.
Bollerslev, T., 1986. Generalized autoregressive conditional heteroskedasticity. J. Econ. 31, 307–327.
Bedendo, M., Campolongo, F., Joossens, E., Saita, F., 2010. Pricing multiasset equity
options: how relevant is the dependence function? Journal of Banking and Finance 34, 788–801.
Bali, T.G., Theodossiou, P., 2007. A conditional-SGT-VaR approach with alternative GARCH models. Ann. Oper. Res. 151 (1), 241–267.
Bali, T.G., Mo, H., Tang, Y., 2008. The role of autoregressive conditional skewness and kurtosis in the estimation of conditional VaR. J. Bank. Finance 32, 269–282.
Bali, T., 2007. A generalized extreme value approach to financial risk measurement.
Journal of Money, Credit and Banking 39, 1613–1649.
Christoffersen, P.F., 1998. Evaluating interval forecasts. International Economic Review 39, 841–862.
Duffie, D., Pan, J., 1997. An overview of value-at-risk. The Journal of Derivatives, Spring, 7–49.
Engle, R.F., 1982. Autoregressive conditional heteroskedasticity with estimates of variance of United Kingdom inflation. Econometrics 50 (4), 987–1007.
Frankel, J.A., Rose, A.K., 1996. Currency crashes in emerging markets: an empirical treatment. Journal of International Economics 41, 351–366.
Fajardo, J., Farias, A., 2010. Derivative pricing using multivariate affine generalized
hyperbolic distributions. Journal of Banking and Finance 34, 1607–1617.
Harris, R.D.F., Kucukozmen, C.C., 2001. The empirical distribution of UK and US stock returns. J. Bus. Finance Acc. 28, 715–740.
Harris, R.D.F., Kucukozmen, C.C., Yilmaz, F., 2004. Skewness in the conditional distribution of daily equity returns. Appl. Finance Econ. 14, 195–202.
Jegadeesh, N., and S. Titman, 1993, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, Journal of Finance, 48, 65-91.
Kaminsky, G.L., Lizondo, S., Reinhart, C.M., 1998. Leading indicators of currency crises. IMF Staff Papers 45 (1).
Kim, Y.S., Rachev, S.T., Bianchi, M.L., Fabozzi, F.J., 2008. Financial market models
with Levy processes and time-varying volatility. Journal of Banking and Finance
32, 1363–1378.
Kupiac, P., 1998. Stress testing in a value at risk framework. The Journal of Derivatives 6 (1) , 7–25.
Kupiec, P.H., 1995. Techniques for verifying the accuracy of risk measurement models. J. Derivatives 3, 73–84.
Hung-Chun Liu, Ming-Chih Lee, Ching-Mo Chang, 2009.The role of SGT distribution in Value-at-Risk estimation: evidence from the WTI crude oil market. Investment Management and Financial Innovations, Volume 6, Issue 1.
McDonald, J., Newey, W., 1988. Partially adaptive estimation of regression models via the generalized t distribution. Econometric Theory 4, 428–457.
Moskowitz, T.J., Grinblatt, M., 1999, Do industries explain momentum? Journal of Finance, 54, 1249-1290.
Sorwar, G., Dowd, K., 2010. Estimating financial risk measures for options. Journal of Banking and Finance 34, 1982–1992.
Theodossiou, P., 1998. Financial data and the skewed generalized t distribution. Manage. Sci. 44 (12), 1650–1661.
Virginie Coudert and Mathieu Gex, 2008, Does risk aversion drive financial crises? Testing the predictive power of empirical indicators, Journal of Empirical Finance 15, 167–184.
Wan-Hsiu Cheng and Jui-Cheng Hung, 2011, Skewness and leptokurtosis in GARCH-typed VaR estimation of petroleum and metal asset returns. Journal of Empirical Finance. 18, 160–173.
Young Shin Kim, Svetlozar T. Rachev, Michele Leonardo Bianchi, Ivan Mitov, and Frank J. Fabozzi, 2011, Time series analysis for financial market meltdowns. Journal of Banking & Finance 35,1879–1891.