本文建立具有厚尾分配型態與基差效果的不對稱GARCH模型，同時探討波動不對 稱行為、資產條件分配與基差等三者對避險績效的影響性。本文應用F檢定與Hansen （2005）的Superior Predictive Ability（SPA）法來檢定避險模型的績效。本文以新加坡 摩根台股股價指數期貨（SIMEX MSCI Taiwan Stock Index Futures, MSCI）為研究對象， 實證結果顯示在避險期間為1天時，常態分配下的GARCH模型與其他模型具有相同的避 險績效，表示波動不對稱性、厚尾分配與基差三者對於1天期的避險績效並無影響效果。 然而在長天期的避險期間，實證結果顯示考慮波動不對稱性、厚尾分配型態與基差能提 升避險的績效，其中以考慮Student t分配與不對稱基差的GARCH模型為最佳。

The purpose of this study is to discuss the asymmetric volatility, basis and conditional fat-tail or skewed distribution which can improve the futures hedging effectiveness of SIMEX MSCI Taiwan Stock Index Futures. This study utilizes the four conditional distributions which are normal distribution, Student t distribution, the generalized error distribution (GED) and the skewed generalized t distribution (SGT), in addition to symmetric and asymmetric basis, also GARCH and GJR models to estimate the assessment of different models to the hedging effectiveness. We perform Hansen (2005) superior predictive ability to test for predictive superiority of our methods over the benchmark model, and find that there is no influence of asymmetric, fat-tail, basis upon the hedging period one day. The findings also find the asymmetric volatility, basis and conditional fat-tail distribution especially the student t distribution and GARCH models of asymmetric basis can improve the hedging effectiveness of stock index futures in long hedging periods.