A comment on a new super-efficiency model in the presence of negative data

Click to increase image sizeClick to decrease image size

The authors of the original paper were contacted but did not wish to provide a Response to this Viewpoint.

Acknowledgements

Adel Hatami-Marbini appreciates the FRS-FNRS for the support he received as a chargé de recherches at the Université catholique de Louvain for this project.

Notes

1 P_io⁻ or P_ro⁺ may be 0 for some outputs or inputs (Portela et al, 2004, p. 1114). Although DMU_o can locate at the ideal point (ie, P_io⁻=P_ro⁺=0) there is no concrete manifestation that any DMU exists at this point in practice. Thus, this paper does not consider a situation with P_io⁻=P_ro⁺=0, analogous to Portela et al (2004).

2 It should be noted that Hadi-Vencheh and Esmaeilzadeh (2013, p.398, Theorem 2) only considers this case.

3 Furthermore, it is important to note that the efficiencies of DMUs {B, K} are also not correctly reported by Hadi-Vencheh and Esmaeilzadeh (2013).