| Article ID: | iaor20083172 |
| Country: | Portugal |
| Volume: | 27 |
| Issue: | 1 |
| Start Page Number: | 21 |
| End Page Number: | 36 |
| Publication Date: | Jun 2007 |
| Journal: | Investigao Operacional |
| Authors: | Gomes Eliane Gonalves, Mello Joo Carlos Correia Baptista Soares de, Meza Lidia Angulo, Fernandes Artur Jos Silva |
| Keywords: | statistics: data envelopment analysis |
In this paper we have two main goals. From a theoretical point of view we intend to develop an improvement to the DEA variable selection models. It is in the variable normalisation step that this improvement takes place. From a practical point of view we intend to measure an energetic efficiency of the 27 Brazilian states. A state achieves this efficiency when it has high values of some socio-economic indexes and low level of energy consumption. We used the non-parametrical approach of Data Envelopment Analysis (CCR and BCC models). We also used the inverted frontier DEA model.