| Summary: | 碩士 === 國立高雄第一科技大學 === 金融系碩士班 === 105 === The aim of this paper is to explore market sentiment and herd behavior among investors and identify relationship among market sentiment, herd behavior, and excess returns for institutional and individual investors. We calculate CAR (Cumulative Abnormal Return) with several controlled factors, including stock sizes, herding measure, and Volatility Index (VIX) which is applied as the proxy of market sentiment in this study. The research was conducted during the period 2010-2013 in the Taiwan stock market. We examine the impact of market sentiment on herd behavior for institutional and individual investors, and the results show that individual investors, who sell stocks intensively when experiencing an upsurge in market fear, behave normally, while institutional investors tend to be calm and not prone to trade in herds with an upswing market sentiment. The study finds that the portfolios that institutional investors herd buy and sell are more profitable than usual, while individual investors tend to lose money when involving trading in herds under low market sentiment. However, the individuals tend to have a better performance when the market is under high sentiment, which could be contributed to the fact that investors have a clear investment direction while the market is relatively anxious. In order to reexamine the relationship between herding measure and the excess return, we replace the herding measure calculated by trade data with herding measure based on order data. Despite the same result for institutional investors, surprisingly, the portfolios made by individual investors under low market sentiment turn to be profitable, which is even more apparent in small stocks. With the results, we can reasonably believe that there are a group of individual investors that are able to gain valuable information and make the right decision on trading, even though they did not trade actively enough to make the deals.
|
|---|
