How Can Long-Run Abnormal Stock Returns Be Both Positively and Negatively Biased?

Posted: 9 May 1998

See all articles by Brad M. Barber

Brad M. Barber

University of California, Davis

John D. Lyon

The University of Melbourne, Department of Accounting and Business Information Systems

Date Written: May 1996

Abstract

We document that long-run market-adjusted cumulative abnormal returns generally yield positively biased test statistics, while long-run market-adjusted buy-and-hold abnormal returns generally yield negatively biased test statistics. However, these general results are sensitive to (1) the period analyzed, (2) the inclusion of NASDAQ firms, and (3) the requirement of pre-event data. These three factors explain the why Barber and Lyon (1996) and Kothari and Warner (1996) obtain apparently contradictory results in their analysis of long-run abnormal returns.

JEL Classification: G12, G14

Suggested Citation

Barber, Brad M. and Lyon, John D., How Can Long-Run Abnormal Stock Returns Be Both Positively and Negatively Biased? (May 1996). Available at SSRN: https://ssrn.com/abstract=7631

Brad M. Barber (Contact Author)

University of California, Davis ( email )

Graduate School of Management
One Shields Avenue
Davis, CA 95616
United States
530-752-0512 (Phone)
530-752-2924 (Fax)

John D. Lyon

The University of Melbourne, Department of Accounting and Business Information Systems ( email )

Victoria
Melbourne, 3010
Australia

HOME PAGE: http://www.abis.unimelb.edu.au/who/staff/john_lyon.html

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
2,620
PlumX Metrics