HAC Standard Errors and the Event Study Methodology: A Cautionary Note

6 Pages Posted: 4 Nov 2008 Last revised: 13 Sep 2010

See all articles by George S. Ford

George S. Ford

Phoenix Center for Advanced Legal & Economic Public Policy Studies

John D. Jackson

Auburn University - Department of Economics

Sarah Skinner

University of Louisiana at Lafayette

Date Written: November 2, 2008

Abstract

In support of Fomby and Murfin (2005), we demonstrate empirically, rather than theoretically, the severe consequences of using HAC standard errors in regression-based financial event studies. Applying an event study to a recent merger, we show that the use of HAC standard errors render misleading conclusions. Critical values for t-tests on the event dummy variables are about 15 times larger than the nominal values using only a year of daily return data. Even with samples of only 100 returns, critical values exceed nominal critical values by a factor of 10.

Keywords: Event Study, HAC, Heteroscedasticity Robust Standard Errors

Suggested Citation

Ford, George S. and Jackson, John Douglas and Skinner, Sarah, HAC Standard Errors and the Event Study Methodology: A Cautionary Note (November 2, 2008). Applied Economics Letters, Vol. 17, No. 12, 2010, Available at SSRN: https://ssrn.com/abstract=1293948

George S. Ford (Contact Author)

Phoenix Center for Advanced Legal & Economic Public Policy Studies ( email )

5335 Wisconsin Avenue, NW
Suite 440
Washington, DC 20015
United States

John Douglas Jackson

Auburn University - Department of Economics ( email )

415 W. Magnolia
Auburn, AL 36849-5242
United States

Sarah Skinner

University of Louisiana at Lafayette ( email )

Lafayette, LA 70504
United States