隨著機構投資人在國內證券市場交易比重的日漸增加，法人的投資行為對國內證券市場的影響將更顯得重要。本研究修正了Lakonishok, Shleifer, and Vishny（1992）提出的從眾行為模型，以2001 年4 月至2004 年3 月為研究期間，衡量國內法人－自營商與國外法人-外資在彼此間以及在不同產業問與不同的投資環境中是否俱有從眾行為上的差異。機構法人是否俱有從眾行為能使我們對價格波動的形成更加暸解，此種同質性的從眾行為對股票市場價格的波動將造成相當程度的影響。實證的結果顯示，外資與自營商在台灣股票市場上存在著顯著的從眾行為，且在各類股中自營商均較外資呈現較明顯的從眾行為。而自營商與外資對不同產業間從眾行為的差異，均一致呈現傳產類股之從眾行為顯著的高於電子與金融類股，但在電子與金融類股之間則無明顯的差異。就不同的漲跌期間而言，外資在各個產業上均顯現出在下跌期間的從眾行為較上漲期間更為顯著，但此現象並末在自營商身上發生。上述的發現說明了機構投資人在資訊愈缺乏與愈悲觀的環境下，其從眾行為會愈明顯。在運用因素分析探討景氣面、籌碼面與風險面因素對從眾行為的影響時，不論外資或自營商之從眾行為均與景氣因素呈現負向的關係'顯現出機構法人對空頭市場的負面消息要較多頭市場的正面消息反應迅速。外資在對電子類股的從眾行為上，除景氣因素外，籌碼因素亦對其從眾行有著負向的關係，意謂著外資在電子類股的投資上，流動性是其相當重視的因素，尤其是在股市下跌的期間。

　　The main focus of this research paper is utilizing the revised LSV model to examine the herding behavior of domestic security dealers and foreign institutional investors in Taiwan's stock market. In addition, we also examine their investment characteristics in different industries under the investing environment from April 2001 to March 2004 Whether institutional investors follow other's trades can help us to understand more about the causes of price movement. This homogeneous herding behavior could be one of the important elements for the fluctuation of the stock price. Empirical finding shows that herding behaviors do exist between the domestic security dealers and the foreign institutional investors in Taiwan's stock market. In any category of stocks, domestic security dealers have much more notably herding behavior when compared to the foreign institutional investors. The different herding behavior of security dealers and foreign institutional investors in different industries shows that herding in traditional industry is much more prominent than electronic industry and financial industry. However, there is no significant different herding pattern between electronic industry and financial industry. In addition, foreign institutional investors display a more obvious herding behavior in bear market than in bull market. The above finding may imply that information and pessimism play an important role in determining of herding behavior for institutional investors. By utilizing factor analysis, we can identify economic prosperity, trading volume and risk factors for explaining the herding behavior of institutional investors. The empirical evidence shows that institutional investors response more quickly for bad news in bear market than good news in bull market. Other than economic factor, trading volume also display a negative effect on herding behavior for foreign institutional investors. This finding indicates that liquidity is also an important factor for foreign institutional investors when investing in the electronic industry, especially in the bear market.