摘要：We apply methodology robust to outliers to an existing eventstudy of the e ect of U.S. nancial reform on the stock markets of the 10largest world economies, and obtain results that di er from the original OLSresults in important ways. This nding underlines the importance of han-dling outliers in event studies. We further review closely the populationof outliers identi ed using Cook's distance and nd that many of the out-liers lie within the event windows. We acknowledge that those data pointslead to inaccurate regression tting; however, we cannot remove them sincethey carry valuable information regarding the event e ect. We study furtherthe residuals of the outliers within event windows and nd that the resid-uals change with application of M-estimators and MM-estimators; in mostcases they became larger, meaning the main prediction equation is pulledback towards the main data population and further from the outliers andindicating more proper tting. We support our empirical results by pseudo-simulation experiments and nd signi cant improvement in determination ofboth types of the event e ect .. abnormal returns and change in systematicrisk. We conclude that robust methods are important for obtaining accuratemeasurement of event e ects in event studies.
