Extended Discriminant Analysis in Data Envelopment Analysis

Data envelopment analysis technique which is developed based on the mathematical programming, evaluates the relative efficiency of a set of homogeneous decision-making units. DA is a method for separating two sets with previous knowledge meanwhile DEA is a technique for separating two sets efficient and inefficient without previous knowledge. Also, goal programming method can be used for both of these methods. DEA-Discriminant Analysis (DEA-DA) that is designed to identify the existence or don’t existence of a overlap between two groups, with using a separation hyper plane and to predict a new sampled observation, when it is belong to one of two groups, because, it does not assume any discriminant function for groups classification. In this article we identify k separation hyper planes for 2k groups, such that, we are doing DA for newly sampled observations in 2k groups by a linear model with k making hyper planes.