Problems related with multicollinearity

Multicolinearity which is defined as the relationship between the explanatory variables in the multiple linear regression model causes the least squares estimator to be a poor estimator. The dr? class estimator which is proposed for this reason is compared to the least squares, ridge and principal components regresssion and the unbiased ridge estimator is compared to the ridge, least squares and class estimators under the matrix mean square error criterion. Due to the effects of multicollinearity on quality control PRESS statistic for the Liu estimator is defined and the estimator of d which minimizes this statistic is given. kr? KeyWords: Multicollinearity, Biased Estimators, Prediction Error Sum of Squares, Quality Control