本研究以1994年到2008年之CISDM避險基金資料為研究對象，並以Fung and Hsieh (2004)的七因子模型為理論基礎，驗證不同環境下避險基金之績效評估。實證結果如下: 1.避險基金指數在經過風險調整後存在顯著的超額報酬；策略分群中，只有「新興市場指數策略」沒有超額報酬，其他避險基金策略都有超額報酬。就系統風險而言，S&P 500指數是相當重要的風險因子；在策略分群中，「債券期貨回望跨式組合報酬率」是僅次於S&P 500指數的風險因子。2. 多空市場下的績效評估中，多頭的超額報酬顯著優於空頭的超額報酬，在策略分群中，多頭市場有13種避險策略存在超額報酬，而空頭市場僅有8種避險策略存在超額報酬。就系統風險而言，S&P 500指數在多空趨勢下都是重要的風險因子。在多頭趨勢中，「商品期貨回望報酬率」是僅次於S&P 500的風險因子；在空頭趨勢中，「穆迪債劵Baa級減掉美國十年期公債月報酬率之差」和「債券期貨回望報酬率」是僅次於S&P 500指數的風險因子。3. 特殊事件下的前後期績效評估，包括東南亞金融風暴、LTCM事件和網路科技 泡沫。總樣本的差異性分析中，東南亞金融風暴和LTCM事件前後皆有超額報酬，事件前小於事件後，但無顯著差異；網路科技泡沫中，事件後的績效比事件前的績效差，但沒顯著差異。 In this thesis, the main database on hedge funds is CISDM from January 1994 to December 2008. We apply Fung and Hsieh's seven-factor model to get alpha value and test the performance under various circumstances. The findings of the paper are as follows: 1.The hedge funds exists the excess returns after controlling risk factors. According to 14 different hedge fund strategies, there is no excess return in Emerging Markets Index. Most important risk factor to exam hedge fund performance is S&P 500 index, and the next is the factor of lookback straddle portfolio return of bond futures.2.The excess return in Bull market is significantly better than in Bear market. We find that there are thirteen investment categories of hedge funds exist alpha value in Bull market, however, there are only eight in Bear market. S&P 500 and credit risk (FR10BAA and PTFSBD) are important factors to exam the hedge funds performance.3.To further evaluate the performance under special events including Asia financial crisis in 1997, LTCM in 2000 and the technology bubbles in 2000. We find that the hedge funds generate alpha value in the period under these events. However, there is not significant difference before and after the occurrence of these events.
