I am studying a new approach for Data Envelopment Analysis. It is basically very similar to the classical approach but with different assumptions. This has a consequence on the final form of the frontier, and of course on the efficiency of the DMU.
I need to make a relatively short comparison between the original and new approach, from a theoretical point of view.
I don't know, however, what would be the points of comparison since the new approach is only a specification of the original.
Is there a generally accepted framework for comparing rigourosly mathematical models?