Order-Value Optimization: Formulation and solution by means of a primal cauchy method

The Order-Value Optimization (OVO) problem is a generalization of the classical Minimax problem. Instead of the maximum of a set functions, the functional value that ranks in the p-th place is minimized. The problem seeks the application to (non-pessimistic) decision making and to model fitting in the presence of (perhaps systematic) outliers. A Cauchy-type method is introduced that solves the problem in the sense that every limit point satisfies an adequate optimality condition. Numerical examples are given.This author was supported by FAPESP (Grant 01/05492-1) and CNPq (Grant 301115/96-6)This author was supported by PRONEX, FAPESP (PT 2001-04597-4)