A new method for increasing the discrimination power of DEA in comparison to PCA

PCA is a method to reduce the dimensionality of data set. On the other hand within DEA, problems of discrimination often arise. So PCA can be applied to DEA in order to increase discrimination. The disadvantage of PCA-DEA is that data must be transformed back to the original form in order to find the targets for improvement. This paper proposes an approach to solve this problem and increase discrimination simultaneously.