Abstract
The superefficiency (SE) model for data envelopment analysis (DEA) measures the influence of each decision-making unit on the efficient frontier and identifies those observations exerting the most influence. As such, the SE model is useful in identifying potential data errors. It also obviates the need for non-Archimedian infinitesimal constraints and eliminates the inherent problem of primal multiple optimal solutions and dual degeneracy characteristic of other DEA model versions.
| Original language | English |
|---|---|
| Pages | 1041-1043 |
| Number of pages | 3 |
| State | Published - 1 Dec 1996 |
| Event | Proceedings of the 1996 27th Annual Meeting of the Decision Sciences Institute. Part 2 (of 3) - Orlando, FL, USA Duration: 24 Nov 1996 → 26 Nov 1996 |
Other
| Other | Proceedings of the 1996 27th Annual Meeting of the Decision Sciences Institute. Part 2 (of 3) |
|---|---|
| City | Orlando, FL, USA |
| Period | 24/11/96 → 26/11/96 |
Fingerprint
Cite this
}
Diagnostic for measuring influence in data envelopment analysis. / O'Neill, Liam; Lewis, Holly S.; Ord, J. Keith.
1996. 1041-1043 Paper presented at Proceedings of the 1996 27th Annual Meeting of the Decision Sciences Institute. Part 2 (of 3), Orlando, FL, USA, .Research output: Contribution to conference › Paper
TY - CONF
T1 - Diagnostic for measuring influence in data envelopment analysis
AU - O'Neill, Liam
AU - Lewis, Holly S.
AU - Ord, J. Keith
PY - 1996/12/1
Y1 - 1996/12/1
N2 - The superefficiency (SE) model for data envelopment analysis (DEA) measures the influence of each decision-making unit on the efficient frontier and identifies those observations exerting the most influence. As such, the SE model is useful in identifying potential data errors. It also obviates the need for non-Archimedian infinitesimal constraints and eliminates the inherent problem of primal multiple optimal solutions and dual degeneracy characteristic of other DEA model versions.
AB - The superefficiency (SE) model for data envelopment analysis (DEA) measures the influence of each decision-making unit on the efficient frontier and identifies those observations exerting the most influence. As such, the SE model is useful in identifying potential data errors. It also obviates the need for non-Archimedian infinitesimal constraints and eliminates the inherent problem of primal multiple optimal solutions and dual degeneracy characteristic of other DEA model versions.
UR - http://www.scopus.com/inward/record.url?scp=0030350358&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:0030350358
SP - 1041
EP - 1043
ER -