Vector approximation problems in geometric vector optimization

Abstract

Using the geometric vector inequality, introduced by Elster/Elster/Go pfert [9], we consider a special case of that inequality which is a basis for treating certain classes of (nonsmooth) vector optimization problems: vector curve fitting problemslp -vector approximation problems and vector regression problems. By the introduced geometric vector inequality we obtain in a natural way dual problems for such special-structured vector optimization problems, which can be solved in an efficient manner, since the dual constraints are linear.