Evaluación mediante Análisis Discriminante de Ratios

Abstract

The purpose of this work is to show a method that provides, for a given group of
variables, common weights for all the units that discriminate optimally between
efficient and inefficient units as previously given by Data Envelopment Analysis
(DEA), and to apply it to rank the Banks of República Argentina. The
Discriminant Analysis of Ratios (DR-DEA) construct a ratio between the
composite outputs and the composite inputs, where the common weights are
computed by a non-linear optimization of the discriminant approach, for the two
groups pregiven by DEA. This method uses as discriminant approach a ratio
function instead of the lineal function used by the traditional Discriminant
Analysis.