LEAST SUM OF ABSOLUTE DEVIATIONS REGRESSION - DISTANCE, LEVERAGE, ANDINFLUENCE

Authors
BERRY KJ MIELKE PW
Citation
Kj. Berry et Pw. Mielke, LEAST SUM OF ABSOLUTE DEVIATIONS REGRESSION - DISTANCE, LEVERAGE, ANDINFLUENCE, Perceptual and motor skills, 86(3), 1998, pp. 1063-1070
Citations number
10
Categorie Soggetti
Psychology, Experimental
Journal title
Perceptual and motor skills → ACNP
ISSN journal
00315125
Volume
86
Issue
3
Year of publication
1998
Part
1
Pages
1063 - 1070
Database
ISI
SICI code
0031-5125(1998)86:3<1063:LSOADR>2.0.ZU;2-2
Abstract
Least sum of absolute deviations regression is considered as a robust alternative to least sum of squared deviations regression. A simple ex ample of linear regression utilizing one predictor variable and a sing le extreme value illustrates the potential effects of distance, levera ge, and influence on the two regression models.