A discrete shrinking method as alternative to least squares
Fikri Öztürk · Communications Faculty Of Science University of Ankara Series A1Mathematics and Statistics · 1984
F or the classical iinear regression problem , a num ber of estim ators alternative to least squares have been proposed for situations in w hich m ulticollinearity is a problem . This paper investigates m ean square error properties of biased regression estim ators and presents a proce- dure for obtaining im proved estim ators.