本文之主要研究目的為檢測波動度期貨的避險績效，並且利用美國地區和歐洲地區不同的波動度期貨加以比較分析。此外，本文利用波動度期貨的避險績效結合大盤股票指數之歷史報酬波動判斷現貨波動之幅度用以判斷交易時機。VIX期貨、Mini-VIX期貨代表的是美國期貨市場的樣本，VSTOXX MINI期貨代表歐洲期貨市場的樣本，現貨部分為S&P500指數、EURO STOXX 50 指數收盤價，樣本期間則涵蓋2010年8月2日至2012年7月31日之每日收盤價交易資料，而實證模型除了採用GARCH外，本文進一步採用不同分配之GARCH模型進行檢測，分別為T分配及厚尾分配之GARCH模型。 實證結果發現：（一）持有期間為一個月時有較佳的避險績效。（二）本文在此利用大盤股票指數報酬的波動性當作判斷標準，發現在美國市場下的避險績效和避險期間呈現正相關，歐洲市場則沒有此現象。（三）實證結果顯示美國市場因為成交量較大，導致市場反應速度較快，故投資人持有期間為短天期時，將會有負的避險績效。（四）期貨市場存在高峰厚尾的特性，實證結果顯示能捕捉厚尾的GARCH-T和GARCH-HT模型較GARCH模型佳。（五）使用DM檢測進行樣本外預測分析，以GARCH-T和GARCH-HT模型進一步分析何者有較佳的預測能力，實證結果為在十天期預測下GARCH-HT模型的預測能力較佳。 This thesis investigates hedge performance of VIX future in American area and Europe area. Besides, the thesis combines stock price index’s history return with hedge performance of VIX future for determining volatility and testing the trading timing. VIX futures, Mini-VIX futures represent sample of the American future market. VSTOXX MINI future represents Europe future madrket. There are S&P500 index and EURO STOXX 50 index in spot market. The thesis uses daily close price data during August 2, 2010 to July 31, 2012. In this paper, three different distributions specified with GARCH model, that is, GARCH, GARCH-T and GARCH-HT. Empirical findings indicate: (1) There is superior hedging effectiveness in one month.(2) This thesis regard stock index return’s volatility as criteria . Hedging effectiveness show positive correlation with hedging period in American area and Europe area don’t. (3) When the investors hold shortly, the hedging effectiveness will be negative in American area. (4) Both GARCH-T model and GARCH-HT model are superior to GARCH model, when the distribution of Future market exhibits leptokurtic and fat-tailed.(5) GARCH-HT model that use DM test to demonstrate the out-sample volatility predicting during in 10 days is more appropriate in comparison with GARCH-T model .
