券商網路推薦資訊實用性之實證研究－以五家券商公司為例

Study on the Information Effect of Website's Recommendations：Evidence of Five Security Firms

陳雅涵(Ya-Han Chen)

指導教授 ： 胡為善

中原大學/商學院/企業管理學系/碩士(2006年)

https://doi.org/10.6840/cycu200600392

摘要

台灣股市中，散戶佔了近八成的比例，而其資訊來源大量仰賴報章雜誌等媒體所推薦的訊息進行判斷，近年來的趨勢顯示，隨著網路科技的發達，愈來愈多的投資人選擇利用網路獲取更快更及時的訊息，然而快則快矣，這些網路推薦訊息真能未投資大眾帶來利益？故本研究選取五家券商---群益、寶來、元大京華、富邦、建華之網路推薦訊息進行分析，以事件研究法之市場模式為主要方法，觀察其專欄推薦是否能為投資人帶來超額報酬，並比較各家推薦效果差異、多家推薦效果，主要結果如下：
（1）全部樣本在公佈日前公佈日當天有顯著之平均異常報酬存在，顯示推薦訊息具有資訊內涵，但值得注意的是，異常報酬從公佈前幾日便已開始反應，並於前一日達到最大；而事件日後平均異常報酬轉而下降，甚至出現負值。若用平均累積異常報酬觀察，若投資人能於公布日前買進推薦個股並於事件日前一日或當天賣出，可獲得正向報酬，然而若投資人選擇當日買進並於推薦日後賣出，所獲之異常報酬反而為負。
（2）受到多家券商推薦個股在公佈日前的表現較單一券商推薦個股為佳，但於事件日後表現較單一券商推薦個股為差。
（3）五家券商均呈現事件日前具顯著正向異報酬，事件日後報酬轉而不顯著、甚至為負的情形，而用ANOVA分析後發現五家券商在推薦日前後的異常報酬無顯著不同，顯示其在資訊外溢、消息掌握度方面，沒有太大差異。以累積異常報酬來看，除非於推薦日前買進持股並於事件日前或當天賣出，否則無法獲得正向報酬。

關鍵字

績效 ； 推薦股票 ； 資訊內涵

並列摘要

During the past three years, there are eighty percent of total investors who traded in Taiwan’s common stock market are private individuals. Their investment information generally comes from newspapers, magazines and so on. As the internet grew rapidly in recent years, there are more and more investors who use internet to search available information for trading stocks. Previous researchers obtained different results on the estimation of stock price trends from newspapers. It motivates this work to examine whether the website’s recommendation on individual company’s common stocks brings benefit to investors or not. This study chooses five large security firms in Taiwan to explore whether their recommendations on individual stocks are accurate or not, and to examine the differences among their recommendation effects. This investigation also compares the difference between stocks which have been recommend by more than one security firm and those have been recommended by only one security firm. The findings are summarized as follows.
(1) All recommended firms had abnormal rate of returns on the event day, suggesting that the information content does exist. However, the abnormal rate of return occurred a few days before the event day and reached the highest on one day before the event day. After the event day, the abnormal rate of return starts to decline, and even turns to negative figures. The CAR shows that if investors bought recommended stocks a few days before the announcement day and sell them on one day before the event day or on the event day, they can gain positive profits, but if they bought on the event day and sold them after the event day, they would not gain profits.
(2) Before the event day, the stocks which had been recommended by two or more security firms outperformed those which had been recommended by only one security firm. However, the findings indicate that after the event day, the result is completely opposite to the results obtained before the event day.
(3) The results also show that the stocks which recommended by all five security firms demonstrating positive abnormal rate of return one day before the event day, and turn to negative figures afterwards. The ANOVA analysis indicates that there are no significant differences among all five security firms. The CAR value also shows that if investors bought the recommended stocks before the announcement day and sold them on one day before the event day or on the event day, they can gain positive profits, but if they bought them on the event day and sold them afterwards, they would not gain profits.

並列關鍵字

information content ； performance ； recommended stocks

參考文獻

Albert, Robert L. Jr. and Timothy R. Smaby, (1996), “Market Response to Analyst Recommendations in the ‘Dartboard’ Column: The Information and Price- pressure Effects”, Review of Financial Economics, Vol. 5, pp. 59-74.

Ball, Ray and Phillip Brown, (1968), “An Empirical Evaluation of Accounting Income Numbers.”, Journal of Accounting Research, Vol. 6, pp. 159-178.

Barber, Brad M. and Douglas Loeffler, (1993), “The 'Dartboard' Column: Second-Hand Information and Price Pressure”, Journal of Financial and Quantitative Analysis, Vol. 28, pp. 273-284.

Bauman,W. Scott, Sudip Datta, and Mai E.Iskandar-Datta,(1995)，“Investment Analyst Recommendations: A Test of ‘The ‘Announcement Effect’ and ‘The Valuable Information Effect ’”, Journal of Business Finance & Accounting , Vol. 22, pp.659-670.

Beja, Avraham. ,(1972),“On Systematic and Unsystematic Components of Financial Risk”, Journal of Finance,pp.37-45.