在衡量投資人情緒方式,目前並無統一方法,而大部分學者都採用市場上的資料來衡量,較少選取直接情緒指標,故本研究將採用兩種不同的衡量方式分別探討與比較,直接情緒指標選用台灣消費者信心指數,而間接情緒指標使用主成分分析法,選取市場週轉率、資券比與三大法人週轉率,建構間接情緒指標,並分別放入Fama-French五因子模型中,探討投資者情緒對於股票報酬之影響,研究發現,當期與前期直接情緒指標化工產業呈現顯著影響;當期間接情緒指標在大部份電子產業呈現負向顯著影響,表示這些電子產業情緒愈高,會削弱報酬,而前期間接情緒指標對股票報酬的預測能力並不好,總結,消費者信心指數所建構的直接情緒指標解釋能力較不佳,而本研究所建構的當期間接情緒指標能解釋近過半的產業報酬。
Now, due to mixed measuring investor sentiment, while most scholars use market data to measure and fewer scholars use direct sentiment indicator. We use both methods to discuss and compare. The direct sentiment indicator uses the Taiwan Consumer Confidence Index. While the indirect sentiment indicator includes the market turnover rate, the ratio of margin/short sale and turnover ratio of three major institutional investors by using principal component analysis (PCA). We add direct and indirect sentiment indicator into Fama-French five-factor model respectively to explore the impact of investor sentiment on stock returns. The empirical results showed that the current and prior direct sentiment indicator has significantly effect in chemical industry. The current indirect sentiment indicator has significantly negative effect in most of electronics industries. The result shows that higher sentiment will cause less returns. The prediction of prior indirect sentiment indicator is not good. In summary, the consumer confidence index is not suitable for measuring investor sentiment, and the current indirect sentiment indicator can explain the majority of industry returns.