TY - JOUR
T1 - Research design issues in grouping-based tests
AU - Lys, Thomas
AU - Sabino, Jowell S.
N1 - Funding Information:
*An earlier draft of this paper was entitled ‘An Analysis of Alternative Research Methods L’sed in Market Ethciency Tests’. Financial support from Deloitte & Touche. KPMG Peat Marwick. and the Accounting Research Center at the J.L. Kellogg Graduate School of Management. North-western University is gratefully acknowledged. We have benefited from helpful comments from Jo) Begley. Victor Bernard. Peggy Bishop, Tim Bollerslev, David Burgstahler. Robert Korajcz>k. Robert klagee. Robert McDonald. Jerold Zimmerman. and participants of the uorkshops at Northwestern University. Univjersity of California-Los Angeles, University of Washington. and University of British Columbia. and especially from comments received from an anonymous referee. All remaining errors are ours.
PY - 1992/12
Y1 - 1992/12
N2 - With grouping, a sample is sorted by an observable variable and the mean values of the dependent variable in the extreme-ranked groups are compared. We show that test power is maximized when the two extreme groups each contain 27% of the sample, a much larger percentage than that typically used in the literature. This result is not sensitive to the distribution of the dependent variable. We also show that regression is unambiguously more powerful than grouping, even when the independent variable is measured with error.
AB - With grouping, a sample is sorted by an observable variable and the mean values of the dependent variable in the extreme-ranked groups are compared. We show that test power is maximized when the two extreme groups each contain 27% of the sample, a much larger percentage than that typically used in the literature. This result is not sensitive to the distribution of the dependent variable. We also show that regression is unambiguously more powerful than grouping, even when the independent variable is measured with error.
UR - http://www.scopus.com/inward/record.url?scp=38249008015&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=38249008015&partnerID=8YFLogxK
U2 - 10.1016/0304-405X(92)90033-T
DO - 10.1016/0304-405X(92)90033-T
M3 - Article
AN - SCOPUS:38249008015
SN - 0304-405X
VL - 32
SP - 355
EP - 387
JO - Journal of Financial Economics
JF - Journal of Financial Economics
IS - 3
ER -