Reexamination of Institutional Herding

碩士 === 國立中興大學 === 財務金融系所 === 101 ===   This paper examines the herding behavior on Sias measurement. We analyze the herding behavior of institutional investors in top three exchanges for all NYSE, AMEX, and NASDAQ stocks. Institutional ownership data are quarterly from March 1999 through September 2...

Full description

Bibliographic Details
Main Authors: Po-Ju Lin, 林伯儒
Other Authors: Sheng-Yung Yang
Format: Others
Language:zh-TW
Published: 2013
Online Access:http://ndltd.ncl.edu.tw/handle/7eeq8d
Description
Summary:碩士 === 國立中興大學 === 財務金融系所 === 101 ===   This paper examines the herding behavior on Sias measurement. We analyze the herding behavior of institutional investors in top three exchanges for all NYSE, AMEX, and NASDAQ stocks. Institutional ownership data are quarterly from March 1999 through September 2012. The cross-sectional correlation can be directly decomposed into the portion that result from following their own trades and the portion that result from following other institutional investor''s trades. Second, we examines the relationship between momentum strategy and herding behavior of institutional investors. Third, we examine how the institutional demand affect the future return of securities. Finally, we analyze the relationship between herding behavior and sentiment indicators.   The empirical result are as follow:the results reveal that institutional investors follow themselves and each other into and out of the same securities over the period of time (herd). Restricting the sample to securities with at least one institutional traders, yields the ratio of cross-sectional correlation from following their own trades are greater than the ratio from following other institutional investor''s trades. The larger and smaller portion of institutional demand are the major part to affect the herding behavior of institutional investors.   We add lag return as a standardized independent variable. The results reveal that institutional momentum trading may result from institutional herding. And institutions'' demand is much more strongly related to their own lag demand than lag returns. No matter what measurement we use, OLS or quantile regression, we find prior-quarter institutional demand is negatively correlated with returns.   Finally, we further explore the association between herding behavior and sentiment indicators. The results reveal that ICS and UMCSENT are positively correlated with "following their own" accounted for the proportion of regression coefficient (statistically significant). That is, when increasing in consumer confidence index, institutional investors will tend to follow their own past trading strategy for investment; While these two indicators are negatively correlated with "following other institution" accounted for the proportion of regression coefficient.