Introduction to Data Envelopment Analysis

Abstract

The agricultural productivity is often based on non-parametric models (DEA), or stochastic models (SFA). In this initial article, the editors start by pointing that the models (DEA and SFA) allow estimating the efficiency of the production frontier and their structural forms. Then, it is presented, in general terms, the differences between DEA and SFA models: DEA model involves the use of technical linear programming to construct a non-parametric piecewise surface, and SFA models comprise econometric models with a random variable, or an error term, including two components: one to account for random effects and another to take care of the technical inefficiency effects. Finally, it shows a comparison between the two approaches (SFA and DEA) and the advantages and disadvantages of their utilizations.