券商投資評等的資訊內涵 – 以台灣市場為例

Abstract

本文主要探討以台灣股票市場券商評估報告的發佈日，建議投資人買進或賣出是否存在異常報酬。利用事件分析法分析，從中計算出在券商發佈研究報告中是否有異常報酬率的存在，同時顯示出券商報告在給予投資人資訊上皆露是否具有完全性，或是有隱含資訊讓投資人可以在券商發布之前從中賺取報酬。全文資料組距時間從2010年至2015年止，年期5年，資料範圍為台灣689家上市櫃公司，分類為傳統產業與電子產業，藉此了解產業的區分不同是否有影響。其次，我們研究券商發佈評價報告中升評與降評兩大類，目的是為瞭解不同評等是否存在異常報酬。

本研究採用16063篇研究報告，進行實證分析，發現升評報告比降評報告在事件期間發生異常報酬為顯著，本文也進一步分析，比較台灣傳統產業與電子產業之異常報酬。

This paper examines the information content of the investment ratings by security firms. By investigating the statistical relation between security firms' rating announcements and abnormal stock return. We use Taiwan stock market as example and attempts to find out any relations between the upgrade ratings, downgrade ratings and individual security firms. This paper uses event study and market model analysis to calculate abnormal return of different ratings by brokerage research reports. The data samples for this paper consist 689 Taiwan public traded companies which include traditional companies and electronics companies. This paper uses about 16063 brokers' research reports collected by Taiwan Economic Journal Database (TEJ+).

Either upgrade or downgrade ratings, our findings show that average abnormal return (AAR) between -1 to 1 day is statistical significant. Regarding the industry sectors, electronics companies have better average abnormal return than traditional companies. Finally, we check the different between target prices and actual stock prices of investment reports to understand brokers' prediction accuracy when it's aforesaid rating.