Pre-test estimation in the linear regression model with competing linear constraints

H. Hessenius & G. Trenkler
This paper introduces and investigates a new pre-test estimator for the parameter vector of the linear regression model. This estimator is based on two sets of linear restrictions - at least one of them consisting of correct information. The statistical properties of the new estimator with respect to its mean square error matrix are investigated and necessary and sufficient conditions showing the potential dominance of this biased estimator over its competitors are derived.