Template-Type: ReDIF-Paper 1.0
Title: On Minimax Estimation in Linear Regression Models with Ellipsoidal Constraints
Author-Name: Norbert Christopeit
Author-Name: Kurt Helmes
Author-Postal: 
Author-Phone: 
Author-Homepage: 
Classification-JEL: 
Keywords: 
Abstract: We consider the simultaneous linear minimax estimation problem in 
 linear models with ellipsoidal constraints imposed on the unknown parameter. 
 Using convex analysis we derive necessary and sufficient optimality 
 conditions for a matrix to define the best linear minimax estimator. For 
 certain regions of the set of characteristics of the linear models and the 
 constraints we exploit these optimality conditions and get explicit formulae 
 for best linear minimax estimators.
Series: Sonderforschungsbereich 303, University of Bonn, Germany
Length: 17 pages
Creation-Date: 1991-10
Revision-Date: 
Handle: RePEc:bon:bonsfb:205