本研究使用2007年1月至2016年2月間分析師盈餘預測資料，以實證分析投資人情緒、分析師盈餘預測誤差對投資組合股價報酬率的影響。實證結果指出在本研究所有期間，當期分析師盈餘預測誤差，不論哪一種投資組合持有一年都具有顯著的正報酬率，但是高與低分析師盈餘預測誤差之投資組合不具統計上顯著的差異性。就不同狀態的投資人期間來看，首先，在投資人情緒高漲期間，分析師盈餘預測誤差僅最低的投資組合在持有一個月皆具有顯著的正報酬率，但是最高與最低分析師盈餘預測誤差之投資組合不具統計上顯著的差異性。其次，在投資人情緒平穩期間，分析師盈餘預測誤差最高投資組合在持有一年都具有顯著的正報酬率。但是高低分析師盈餘預測誤差之投資組合不具統計上顯著的差異性；最後，在投資人情緒低落期間，分析師盈餘預測誤差不論哪一種投資組合持有9個月都具有顯著的正報酬率，但是最高與最低分析師盈餘預測誤差之投資組合不具統計上顯著的差異性。 Using the data on security analyst for earning forecast over January 2007 to February 2016, this thesis empirically explores the impacts of investor’s sentiment and earning forecast errors on stock returns based on portfolio analysis. Empirical results indicate that during the all sample period current earning forecast errors have significant impacts on stock returns for all ranking combinations, but there is no significant difference in the high and low earning forecast errors. Regarding the different states of investor’s sentiment, during the period of higher investor’s sentiment, the lowest portfolios sorted by earning forecast errors show significant and positive returns. Moreover, during the period of moderate investor’s sentiment, the lowest portfolios sorted by earning forecast errors also present significant and positive returns. Finally, during the period of lower investor’s sentiment, the lowest portfolios sorted by earning forecast errors show significant and positive returns over nine months. However, differences between high and low sorted by earning forecast errors for all portfolios exhibit insignificance for the stock returns.
