Some comments on Statistical Papers 46, 270–395 (2005)

Abstract

In Khan et al. (2005) three biased estimators for the intercept of a linear regression model are discussed. The estimators are based on the maximum likelihood estimator (MLE) and use additional uncertain prior information. The mean square errors (MSE) of the biased estimators are compared with theMSE ofMLE and the authors conclude that for small bias the proposed estimators perform better than MLE and other biased and unbiased estimators.