The PMP methodology is extended to deal with many economic units, self-selection arising from heterogeneous economic behavior, and multiple limiting inputs. The novel structure for the analysis of this type of problems acquires the name “symmetric positive equilibrium problem.” This methodology is articulated in three phases that parallel those one of positive mathematical programming. The recovery of a cost function that calibrates the base-year decisions of each economic unit is realized within a maximum entropy framework. The methodology is illustrated by the analysis of a sample of Italian farms that operate under the complex policy mandates of the European Union. Copyright 2001, Oxford University Press.