本研究在探討以台灣50指數成分個股的財務因子做為變數,嘗試找出打敗大盤績效之投資組合。資料選取期間為2009年1月至2017年12月。前54個月透過統計方法迴歸分析找出顯著效果之財務因子做為變數,並以此顯著效果之財務因子於後54個月使用最小平方法做預測,找出報酬排名前20%共十檔個股做為投資組合。我們將以此投資組合與台灣50進行績效比較,比較期間為2014年至2017年,採用定期抽樣觀察與配對T檢定,實證結果如下: 1. 績效打敗台灣五十(0050)的持有期間,多集中在持有一個月、六個月、九個月;持有一年以上績效打敗台灣五十的次數明顯下滑。 2. 在2014年至2017四個年度中,績效打敗台灣五十的次數以2014年與2017年次數較多,分別為2014年的17次與2017年8次。 3. 在台灣五十(0050)為多頭格局時,T10(前20%的投資組合)較易取得超額報酬;在台灣五十(0050)為空頭格局時,T10(前20%的投資組合)沒有台灣五十(0050)抗跌。 4. 在台灣五十(0050)修正接近尾聲時進場,一年內取得超額報酬的機會較高。有明顯的擇時效果。 ;Abstract In this study, we use the financial factors of the Taiwan 50 Index as the variables, trying to build a portfolio to beat the market. The data selection period is from January 2009 to December 2017. In the first 54 months, regression analysis was used to identify the significant financial factors as variables, and the financial factors with this significant effect were used to predict in the next 54 months to find out the top 20% of 10 stocks for the portfolio. We’ll compare the performance of this portfolio with that of Taiwan 50. From 2014 to 2017, we used regular sampling observations and paired T tests methods. The empirical results are as follows: 1. The performance of defeating Taiwan 50 (0050) was concentrated in holding one month, six months, and nine months; when holding time more than one year, the times of beating Taiwan 50 was obviously declined. 2. From 2014 to 2017, the number of times that performance beat Taiwan 50 was more in 2014 and 2017, 17 times in 2014 and 8 times in 2017 individually. 3. When Taiwan 50 (0050) is in Bull market, T10 (the top 20% of the portfolio) can obtain excess returns easily; when Taiwan 50 (0050) is in Bear market, T10 (the top 20% of the portfolio) cannot defense price fall compare with Taiwan 50. 4. When Taiwan 50 (0050) is approaching the end of collapse, it will be a better chance to get excess return within one year. There is a clear timing effect.