本論文提出在95%、99%與99.5%信賴水準下，GARCH模型建立在SGT分配的風險值估計。在計算條件SGT分配風險值方法上是以GARCH-N模型、GARCH-SGT模型、EGARCH-SGT模型和GJR-SGT模型等四種GARCH模型來配適適合度。樣本是採用能源商品(天然氣)、農產品(小麥)和金屬商品(黃金)之每日期貨結算價。此外本論文在文獻回顧將最近十年在期刊上所蒐集到的風險值論文分類成能源市場、商品市場、股票市場、外匯市場、利率市場和混合型財務市場，使讀者可以知道大部份學者較關心在哪部份。實證結果隱含本文採用的三種樣本對於正面消息產生時相對於負面消息產生會影響波動較大，並且實際上會導致波動度增加。並且本結果顯示在95%信賴水準時， GJR-SGT模型、 EGARCH-SGT模型與GARCH-SGT模型分別對天然氣、小麥與黃金有最佳績效表現。在99%信賴水準時，GARCH-SGT模型對天然氣表現出較精確的風險值估計，然而EGARCH-SGT模型在小麥與黃金部分表現較好。在99.5%信賴水準時，GJR-SGT模型在天然氣與黃金方面績效表現勝過其他模型，然而小麥在GARCH-N模型表現較好。此結果隱含具偏態、峰態與厚尾特性之SGT分配應用於風險值估計時，其效果優於常態分配。因此，可以應用在此極端事件下的商品市場風險管理上，使風險值較不會誤差過大而低估，並且有助於投資人更精確地評估在這三種商品面對此財務危機與呈現更大波動下可能的最大損失。綜觀在全球金融海嘯下，本研究結果提供欲從事商品市場風險管理的投資大眾，以GARCH類模型搭配SGT分配的風險值估計，仍是有效之市場風險管理工具的有利證據。

This thesis proposes the estimation of model-based VaR measures based on the skewed generalized t (SGT) distribution at three confidence levels (95%, 99%, and 99.5%). The suitability of four GARCH models (GARCH-N, GARCH-SGT, EGARCH-SGT, and GJR-SGT) in computing conditional-SGT-VaR measures is addressed. The daily futures price data for a collection of commodities energy (Natural Gas), agricultural (wheat) and spanning metal (gold), are used. In addition, this thesis presents a concise summary of the findings of the selected studies in last decade that are divided into six categories (energy market, commodity market, stock market, foreign exchange rate market, interest rate market, and various financial market), we can find out which category most researchers concentrate in. The empirical results imply that the positive news has a greater effect on volatility and a positive return shock actually increases volatility for three cases during the Global Financial Turmoil. Moreover, the results show that at the 95% confidence level, the GJR-SGT model, the EGARCH-SGT model and the GARCH-SGT model has the best overall performance for natural gas, wheat and gold, respectively. At the 99% confidence level, the GARCH-SGT (the EGARCH-SGT) model provides more accurate VaR estimates for natural gas (wheat and gold). At the 99.5% confidence level, the GJR-SGT model (the GARCH-N model) outperforms the other models for natural gas and gold (wheat). Overall, the SGT error with skewness, kurtosis and fat tails is much superior to the normal error in capturing the downside risk for most commodity cases. The finding can be apply in risk management to capturing the extreme event in commodity markets, in this manner, the VaRs’ errors will not be overlarge and underestimate the VaRs, and it will be helpful that investors evaluate more precisely the possible maximum losses when the three commodities face the Global Financial Turmoil and become more volatile. These results confirm that Value-at-Risk-based risk management using the GARCH-type models with SGT distribution are still valid for commodity markets during the Global Financial Turmoil.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS                                               i                        
ABSTRACT IN CHINESE                                               ii
ABSTRACT IN ENGLISH                                               iv
LIST OF TABLES                                                     viii
LIST OF FIGURES													  ix

CHAPTER
1.Introduction 1
1.1 Motivations and objectives 1
1.1.1 The expatiation of Value-at-Risk environment 1
1.1.2 The background of subprime mortgage storm 4
1.1.3 The reasons of selecting commodity market 5
1.1.4 The theme and aspect of this thesis 8
1.2 Flow chart 9
2.Literature Review 10
2.1 Energy market	10
2.2 Commodity market 12
2.3 Stock market 13
2.4 Foreign exchange rate market 20
2.5 Interest rate market 22
2.6 Various financial markets 23
3.Econometric Methodology 32
3.1 The definition of Value-at-Risk	32
3.2 Various GARCH genre of volatility models	33
3.2.1 GARCH (1, 1) model 34
3.2.2 GJR-GARCH model 35
3.2.3 EGARCH model 36
3.3 The Skewed Generalized T (SGT) distribution 37
3.4 Calculation of the model-based VaR 39
3.5 Performance evaluation of Value-at-Risk models 40
3.5.1 Binary loss function (BLF) 40
3.5.2 Regulatory loss function (RLF)	41
3.5.3 LR test for unconditional coverage (LRuc ) 41
3.5.4 LR test for conditional coverage (LRcc)	42
4.Empirical Results and Analysis 44
4.1 Data description and preliminary analysis	44
4.2 Model estimates and diagnostic test 49
4.3 Analyzing predictive accuracy of the model-based VaR 52
4.3.1 VaR performance for various models at the 95% confidence level 53
4.3.2 VaR performance for various models at the 99% confidence level 56
4.3.3 VaR performance for various models at the 99.5% confidence level 59
4.4 Graph identification and analysis 62
5.Conclusions and Suggestions 67
References 70

LIST OF TABLES
Table 2.1 Summary of the findings of the selected studies in last decade 26
Table 4.1 The futures contract specifications of various commodities used in this thesis 44
Table 4.2 Descriptive statistics of daily returns using futures data 48
Table 4.3 Estimation results of various GARCH models 51
Table 4.4 Summary results of various model-based VaRs at the 95% confidence level 55
Table 4.5 Summary results of various model-based VaRs at the 99% confidence level 58
Table 4.6 Summary results of various model-based VaRs at the 99.5% confidence level	61

LIST OF FIGURES

Figure 4.1 Daily returns and alternative model-based VaR forecast for Natural Gas 64
Figure 4.2 Daily returns and alternative model-based VaR forecast for Wheat 65
Figure 4.3 Daily returns and alternative model-based VaR forecast for Gold 66

[1].Ané, T., (2006), “An analysis of the flexibility of asymmetric power GARCH models”. Computational Statistics & Data Analysis, 51, 1293-1311.
[2].Angelidis, T., Benos, A., and Degiannakis, S., (2004), “The use of GARCH models in VaR estimation”. Statistical Methodology, 1, 105-128. 
[3].Angelidis, T. and Degiannakis, S., (2005), “Modeling risk for long and short trading positions”. Journal of Risk Finance, 6, 226-238. 
[4].Almeida C. and Vicente, J., (2009), “Are interest rate options important for the assessment of interest rate risk?”. Journal of Banking & Finance (in press). 
[5].Bali, T.G. and Theodossiou, P., (2007), “A conditional-SGT-VaR approach with alternative GARCH models”. Annals of Operations Research, 151, 241-267.
[6].Bams, D., Lehnert, T., and Wolff, C.C.P., (2005), “An evaluation framework for alternative VaR-models”. Journal of International Money and Finance, 24, 944-958. 
[7].Bollerslev, T., (1986), “Generalized autoregressive conditional heteroscedasticity”. Journal of Econometrics, 31, 307–327.
[8].Christoffersen, P.F. (1998), “Evaluating interval forecasts”. International Economic Review, 39, 841–862.
[9].Costello, A., Asem, E., and Gardner, E., (2008), “Comparison of historically simulated VaR: Evidence from oil prices”. Energy Economics, 30, 2154-2166. 
[10].Cotter, J., (2008), “Varying the VaR for unconditional and conditional environments”. Journal of International Money and Finance, 26, 1338-1354.
[11].Da Veiga, B., Chan F., and McAleer, M., (2008), “Evaluating the impact of market reforms on Value-at-Risk forecasts of Chinese A and B shares”. Pacific-Basin Finance Journal, 4, 453-475. 
[12].Dowd, K., (1998), “Beyond Value at Risk: the new science of risk management”. John Wiley & Sons Ltd.  
[13].Engle, R.F., (1982), “Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation”. Econometrica, 50, 987–1007.
[14].Fan, Y., Zhang, Y.J., Tsai, H.T., and Wei, Y.M., (2008), “Estimating Value at Risk of crude oil price and its spillover effect using the GED-GARCH approach”. Energy Economics, 30, 3156-3171.
[15].Giot, P., (2003), “The information content of implied volatility in agricultural commodity markets”. Journal of Futures Markets, 23, 441–454.
[16].Giot, P. and Laurent, S., (2003a), “Market risk in commodity markets: A VaR approach”. Energy Economics, 25, 435-457. 
[17].Giot, P. and Laurent, S., (2003b), “Value-at-Risk for long and short trading positions”. Journal of Applied Econometrics, 18, 641-664. 
[18].Glosten, L.R., Jagannathan, R., and Runkle, D.E., (1993), “On the relation between the expected value and the volatility of the nominal excess return on stocks”. Journal of Finance, 48, 1779–1801.
[19].Hansen, B.E., (1994), “Autoregressive conditional density estimation”. International Economic Review, 35, 705–730.
[20].Huang, Y. C. and Lin, B. J., (2004), “Value-at-Risk analysis for Taiwan stock index futures: Fat tails and conditional asymmetries in return innovations”. Review of Quantitative Finance and Accounting, 22, 79-95. 
[21].Hung, J.C., Lee, M.C., and Liu, H.C., (2008), “Estimation of Value-at-Risk for energy commodities via fat-tailed GARCH models”. Energy Economics, 30, 1173-1191.
[22].Jondeau, E. and Rockinger, M., (2003), “Conditional volatility, skewness, and kurtosis: existence, persistence, and comovements”. Journal of Economic Dynamics and Control, 27(10), 1699-1737.
[23].Jorion, P., (2000), “Value at Risk：The new benchmark for controlling market risk”. McGraw-Hill, Inc, 2.
[24].Jorion, P., (2001), “Value at Risk — The new benchmark for managing financial   risk”. The McGraw-Hill Co.
[25].Kupiec, P., (1995), “Techniques for verifying the accuracy of risk management models”. Journal of Derivatives, 3, 73-84.
[26].Lee, M.C., Su, J.B., and Liu, H.C., (2008), “Value-at-Risk in US stock indices with skewed generalized error distribution”. Applied Financial Economics Letters, 4(6), 425-431. 
[27].Ling, S. and McAleer, M., (2002), “Necessary and sufficient moment conditions for the GARCH(r,s) and asymmetric power GARCH(r,s) models”. Econometric Theory, 18, 722-729.
[28].Lopez, J.A., (1998), “Methods for evaluating Value-at-Risk estimates”. Economic Policy Review, 119–124.
[29].Marimoutou, V., Raggad, B., and Trabelsi, A., (2009), “Extreme value theory and Value at Risk: Application to oil market”. Energy Economics, 12, 1741-1752.
[30].Mazin A.M. and Al Janabi, (2006), “Foreign-exchange trading risk management with value at risk Case analysis of the Moroccan market”. Journal of Risk Finance, 7(3), 273-291. 
[31].McDonald, J.B. and Newey, W.K., (1988), “Partially adaptive estimation of regression models via the generalized t distribution”. Econometric Theory, 4, 428–457.
[32].McMillan, D.G., Speight, A.E.H., and Evans, K.P., (2008), “How useful is intraday data for evaluating daily Value-at-Risk? : Evidence from three Euro rates”. Journal of Multinational Financial Management, 18, 488-503.
[33].Nelson, D., (1991), “Conditional heteroskedasticity in asset returns: A new approach”. Econometrica, 59, 347–370.
[34].Pascual, L., Romo, J., and Ruiz, E., (2005), “Bootstrap prediction for returns and volatilities in GARCH models”. Computational Statistics & Data Analysis, 50(9), 2293-2312.[35].So, M.K.P. and Yu, P.L.H., (2006), “Empirical analysis of GARCH models in value at risk estimation”. International Financial Markets, Institutions & Money, 16, 180-197.
[36].Su, E. and Knowles, T.W., (2006), “Asian pacific stock market volatility modeling and Value at Risk analysis”. Emerging Markets Finance and Trade, 42, 18-62. 
[37].Subbotin, M.T.H., (1923), “On the law of frequency of error”. Matematicheskii Sbornik, 31, 296–301.
[38].Theodossiou, P., (1998), “Financial data and the skewed generalized t distribution”. Management Science, 44, 1650–1661.
[39].Wu, P.T. and Shieh, S.J., (2007), “Value-at-Risk analysis for long-term interest rate futures: Fat-tail and long memory in return innovations”. Journal of Empirical Finance, 14, 248–259.
[40].Xekalaki, E. and Degiannakis, S., (2005), “Evaluating volatility forecasts in option pricing in the context of a simulated options market”. Computat. Statist. Data Anal, 49 (2), 611–629.