PRIMAL-DUAL APPROACH TO NONPARAMETRIC PRODUCTIVITY ANALYSIS - THE CASE OF US AGRICULTURE

Nonparametric methods for measuring productivity indexes based on boun ds for the underlying production technology are presented. Following B anker and Maindiratta, the lower bound is obtained from a primal appro ach while the upper bound corresponds to a dual approach to nonparamet ric production analysis. These nonparametric bounds are then used to e stimate input-based and output-based distance functions. These radial measures provide the basis for measuring productivity indexes. Applica tion to times series data on U.S. agriculture indicates a large gap be tween the primal lower bound and the dual upper bound. This generates striking differences between the primal and dual nonparametric product ivity indexes.