On the Use of a Non-Singular Linear Transformation of Variables in Data Envelopment Analysis

In this paper, we consider non-singular linear transformation of the inputand outputvariables in the Data Envelopment Analysis (DEA). The transformation is useful in selecting variables and dealing, for instance, with interval scale variables. We will develop general theory and show that the results are invariant due to non-singular linear transformation provided the concept of “dominance” is defined accordingly. The invariance property is valid only for non-singular linear transformation. Finally, we briefly discuss in singular linear transformation and illustrate some pitfalls, which may lead to wrong results.