本研究透過事件研究法(Event Study),以市場模式(Market Model)探討券商發佈之投資評等的資訊內涵及其對股票異常報酬之影響。針對投資評等進行四階段的分析,第一階段:全體券商發佈之投資評等之資訊內涵;第二階段:券商投資評等語言之資訊內涵;第三階段:本國及外資券商所發佈的投資評等之資訊內涵差異;第四階段:本國及外資券商發佈之不同等級投資評等的資訊內涵差異。研究結果顯示:(一)券商評等以正面評等為主,占所有評等的半數,本國券商偏向避免發佈負面評等,其負面評等數量僅占其所有評等的2.3%;(二)中立評等在投資人眼中並不中立,當發佈中立評等時,其異常報酬變化與負面評等趨勢類似,唯在程度上比起負面評等緩和;(三)券商發佈正面評等時是異常報酬率上昇幅度開始趨緩之時:無論是外資或是本國券商都在發佈正面投資評等前四天就出現異常報酬率,發佈日後異常報酬反而從發佈前的大幅上昇轉變為緩步上昇;(四)當外資券商發佈負面評等時,在發佈日異常報酬率會略為下降,但第二日後恢復發佈前水準,反觀本土券商的異常報酬則在發佈後一路下滑。
We examine how stock recommendations relate to abnormal return (AR). Through event study, we used market model to estimate expected returns on each stock, and observed how AR changes before and after the issue dates of stock recommendations. There are four stages of analysis including: (1) How stock recommendations relate to AR; (2) How different level of stock recommendations relates to AR; (3) How different issuers of stock recommendations relate to AR; (4) How different issuers and level of stock recommendation relate to AR. The results indicate that: (1) Stock recommendations tend to be positive, more than half of the total are positive. The local stock recommendations issuers’ negative recommendations contribute only 2.3% of total. (2) Neutral is perceived as negative, the AR shows the same pattern, but smaller in scale. (3) The positive recommendation does not relate to massive growth in AR. AR shows a greater growth before event date than after. (4) After the negative recommendations issued, those come from local issuers relate to a continuous decline in AR, while those come foreign issuers relate to a decline in event date only.