台灣股市農曆月份效應之研究

An Analysis of the Lunar Calendar Monthly Effects on the Taiwan Stock Market

元月效應是股票市場中每年元月報酬會高於其他月份之異常現象。而在亞洲，多數國家傳統上使用農民曆的國家，其市場可能存有特殊的『春節效應』。本研究以台灣股市為對象，檢驗我國股票集中市場是否存有類似之異常效應。研究期間自1995至2014年，共計二十年。研究範圍除了台灣股市之大盤外，將針對容易產生這些異常現象的產業——電子類、金融類、塑膠類、食品類和汽車類指數逐一分析，以提供投資者在股票投資時更多資訊。
'實証結果如下：1.台灣股市在1995至2014年期間，除了存在著農曆月份效應外，亦存在春節效應。加入控制變數後發現，大盤指數與S&P 500指數、美金兌台幣匯率和M1B貨幣供給量有很高的關聯性，並且使得大盤指數之農曆月份效應更為明顯。2.橫斷面分析結果，在電子類、金融類、塑膠類、食品類和汽車類指數部分，除電子類、塑膠類指數有春節效應外，其餘的農曆月份效應並不明顯。3.縱斷面分析結果，透過將20年研究時間區分為四個子期間，發現台灣股市大盤指數報酬之春節效應已經逐漸消失，其餘的農曆月份效應雖然還存在但也產生變化，從最早的農曆三月、十二月轉至九月，近五年來甚至已經逐漸消失。

January Effect' means that the stock market returns in January are generally higher than the returns of all other months. However, many countries in Asia such as China, Japan Korea and Taiwan, the traditional Lunar Calendar is still widely used by general people in their daily life by now. Therefore in those countries the January effects could be quite different patterns and cycles, say 'Lunar January Effect'. Using data of recent twenty years, this study examines the Lunar January effect on the Taiwan stock market.
Empirical results obtained are summarized as follows. 1. During the past twenty years, Taiwan stock market found existed Lunar January effect and the Lunar monthly effects. 2. Cross-sectional analysis shows that Electronics and Plastic sector indexes demonstrated Lunar monthly effects, whereas Financial, Food and Automotive sector indexes shown no such effect. 3. Time series analysis shows that across the past twenty years, the patterns of Lunar monthly effect have shift form March and December in the early periods into September in the later periods. In addition, Lunar January effect has now become much less intensive or barely a disappeared phenomenon in Taiwan.