On the Use of Some Selected Estimators in the Computation of Interactions in a Moderated Multiple Regression of a Masked Survey Data

Okagbue, H. I. and Adamu, M. O. and Edeki, S.O. and Opanuga, A. A. (2016) On the Use of Some Selected Estimators in the Computation of Interactions in a Moderated Multiple Regression of a Masked Survey Data. International Business Management, 10 (4). pp. 352-356. ISSN 1993-5250

PDF Download (122kB)

Abstract

Mean has been traditionally used in the computation of the interactions in moderated linear regression. This research extended to the use of medilllll, mode, trimmed mean and some M-estimators which gave almost the same results when compared with mean. The values ofR2 , MSR and F were the same for all the estimators. Also, simulated results vary slightly with the theoretical results due to the effects of outliers as a result of random mnnbers. The smvey data was masked because it is a subset of an ongoing research work

Item Type:	Article
Uncontrolled Keywords:	Estimators, regression, moderation, mean, MSR
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Engineering, Science and Mathematics > School of Mathematics
Depositing User:	Mrs Patricia Nwokealisi
Date Deposited:	11 Apr 2016 12:13
Last Modified:	11 Apr 2016 12:13
URI:	http://eprints.covenantuniversity.edu.ng/id/eprint/6543

Actions (login required)

View Item