TAIEX避險組合的風險估計：GARCH-EVT-COPULA模型的應用

Estimating Risk of TAIEX Hedging Portfolio:Application of GARCH-EVT-COPULA Model

戴于庭(Yu-Ting Tai)

指導教授 ： 莊忠柱

共同指導教授 ： 李達期(Ta-Chi Lee)

淡江大學/商管學院/管理科學學系碩士班/碩士(2018年)

https://doi.org/10.6846/TKU.2018.00823

摘要

金融資產報酬序列的波動性叢聚、厚尾與極端事件常產生資產報酬尾部行為相依結構的改變，因而借助適當計量模型建構避險組合，可增加投資風險管理效益。本研究以2001年1月2日至2017年12月29日的臺灣加權股價指數現貨每日收盤價與指數期貨的價格為研究對象。在移動視窗法(Rolling-Window)架構下，利用GARCH-EVT-COPULA模型，探討Normal分配與學生t分配下，臺灣加權股價指數與指數期貨的最小變異數避險組合之動態風險管理績效。本研究發現學生t分配的GARCH-EVT-COPULA模型比Normal分配的GARCH-EVT-COPULA模型更能捕捉尾部行為且提高風險管理效益。此外，本研究發現319槍擊案與福島核事件發生後的學生t分配的動態風險管理績效較Normal分配準確。另外，金融海嘯發生後的Normal分配與學生t分配的最小變異數避險組合的預期不足額有顯著差異。本研究的研究成果可提供投資人參考。

關鍵字

期貨 ； GJR-GARCH ； COPULA ； 極值理論 ； 蒙地卡羅模擬法 ； 風險值 ； 預期不足額

並列摘要

The volatility clustering, fat-tail due to rare events, and the dependence structure of tail between of financial asset returns time series may change. Hence, the construction of a minimum variance hedging portfolio(MVHP) from a proper econometric model can increase the effectiveness of risk management. The study examined Taiwan index spot daily close price and Taiwan index futures transaction price occurred close to 13:30 from January 2, 2001 to December 22, 2017. The rolling-window framework and GARCH-EVT-COPULA model are used to measure Value-at-Risk and expected shortfall of the MVHP for the effectiveness of dynamic risk management. The effectiveness of risk management is also compared between Normal distribution and Student t distribution on GARCH-EVT-COPULA model.
　　The empirical results show that the model of Student t distribution is better than the one of Normal distribution. Moreover, results also show that 319-shooting incident and Japan's Fukushima nuclear explosion are compared between GARCH-EVT-COPULA model with Normal distribution and Student t distribution. In addition, the GARCH-EVT-COPULA model with Student t distribution is different from the one with Normal distribution after financial crisis. The evidences have direct implications for investors and risk managers during extreme index futures market comovements.

並列關鍵字

futures ； GJR-GARCH-EVT-COPULA ； VaR ； Monte Carlo Simulation ； Expected Shortfall

參考文獻

1. Ab Razak, R., and Ismail, N. (2016). Portfolio risks of bivariate financial returns using Copula-VaR approach: A case study on Malaysia and US stock markets. Global Journal of Pure and Applied Mathematics, 12(3), 1947-1964

2. Acatrinei, M. (2015). A Copula-GARCH model for a proxy portfolio for bet-fi index. Studii Financiare (Financial Studies), 19(2), 8-16.

3. Acerbi, C., and Tasche, D. (2002). On the coherence of expected shortfall. Journal of Banking & Finance, 26(7), 1487-1503.

4. Aloui, C., and Jammazi, R. (2015). Dependence and risk assessment for oil prices and exchange rate portfolios: A wavelet based approach. Physica A: Statistical Mechanics and its Applications, 436, 62-86.

5. Álvarez-Díez, S., Alfaro-Cid, E., and Fernández-Blanco, M. O. (2016). Hedging foreign exchange rate risk: multi-currency diversification. European Journal of Management and Business Economics, 25(1), 2-7.