動態價格跳躍與最小變異數避險組合的風險值－以英國布蘭特石油現貨與期貨價格為例

Dynamic Price Jump and Value-at-Risk for the Minimum Variance Hedging Portfolio: The Case of Brent Crude Oil Spot and Futures Price

王植顥(Chih-Hao Wang)

指導教授 ： 莊忠柱

共同指導教授 ： 王譯賢(Yi-Hsien Wang)

淡江大學/管理學院/管理科學學系碩士班/碩士(2012年)

https://doi.org/10.6846/TKU.2012.00707

摘要

投資產品的多元化，如何降低投資組合風險是件非常重要的課題。由於期貨市場與現貨市場具有高度相關，利用期貨契約規避市場現貨價格變動的衝擊，已成為投資者維持損益的關鍵。本研究以2010年至2011年布蘭特石油價格為主要研究對象利用Chan and Young(2006)提出的雙變量ARJI-GARCH(1,1)模型捕捉價格不連續的變動，且在95%及99%信賴水準下，比較未避險模型、雙變量GARCH(1,1)模型與雙變量ARJI- GARCH(1,1)模型的最小變異數避險組合的多頭部位風險值。將風險值和資產真實損益間進行比較，最後利用Kupiec(1995)概似比檢定與Christoffersen(1998)條件涵蓋檢定法評估風險值模型。研究結果發現：對風險值績效評估而言，未避險模型與雙變量GARCH(1,1)模型在顯著水準1%下的風險值績效皆未能通過回顧測試，而雙變量ARJI- GARCH(1,1)模型在顯著水準5%及1%下皆呈現顯著，表示此雙變量ARJI-GARCH (1,1)模型能確切地捕捉到厚尾及價格不連續的特性，此結果可為金融機構來評估資產組合下的風險，其能改善在尾部分佈下的風險值估計效能，並捕捉金融資產報酬的波動性叢聚、厚尾及價格不連續等特性。

關鍵字

期貨 ； 風險值 ； 最小變異數避險組合 ； 回顧測試 ； 穿透次數

並列摘要

How to reduce portfolio risk an very important issue based on diversification of investment products. A highly relationship was shown on futures and spot market, so making use of futures contracts to avoid the spot market changes in price shocks has become the key to investors to maintain the profit and loss. The study is to compute Value-at-Risk in the long trading position for the minimum variance hedging portfolio by using GARCH(1,1) model, bivariate GARCH (1,1) model and bivariate ARJI-GARCH(1,1) model at two significant levels. The sample daily data is Brent crude oil closing price in spot and furtures market. Moreover, the study is to evaluate the different models by using backtesting method based on likelihood ratio test proposed by Kupiec (1995) and Christoffersen(1998). The study showed that the bivariate ARJI-GARCH (1,1) model at the significance level of 5% and 1% are statistically significant, indicating that the bivariate ARJI-GARCH(1,1) model can accurately describe the discontinuous characteristics of the fat tail and price. This result can provide valuable information for financial institutions to assess the portfolio risk. It improves the estimated performace under the tail distribution and further forecast the financial asset return volatility, fat tail, and the discontinuous price.

並列關鍵字

Futures ； Value-at-Risk ； Minimum Variance Hedging Portfolio ； Backtesting ； Failure Times

參考文獻

3.黃聖志、蘇欣玟與杜國賓，2008。避險基金指數之風險值探討，商管科技季刊，第九卷，第三期，頁277-300。

4.劉洪鈞、黃聖志與王怡文 (2008)。西德州與布蘭特原油避險策略，真理財經學報，第十八期，頁71-98。

2.胥愛琦 與 吳清豐，2003。台灣股市報酬與匯率變動之波動性外溢效果－雙變量EGARCH模型的應用，台灣金融財務季刊，第四卷，第三期，頁87-103。

5.蘇榮斌，2010，美國原油市場之風險值預測，中華科技大學學報，第四十二期，頁161-175。

1.Angelidis, T., A. Benos and S. Degiannakis, 2007, A Robust VaR Model under Different Time Periods and Weighting Schemes, Review of Quantitative Finance and Accounting, 28, 2, pp. 187-201.