A NEW RESTRICTED MODEL USINGCORRELATION COEFFICIENTS AS AN ALTERNATIVE TO CROSS-EFFICIENCY EVALUATION IN DATA ENVELOPMENT ANALYSIS

In this paper, a new Data Envelopment Analysis (DEA) model, which is restricted using correlation coefficients (the CCRCOR), was compared with the cross-efficiency evaluation model (the CEM) and some other models. We used two hypothetical data sets, which were generated by E. Thanassoulis (A comparison of regression analysis and data envelopment analysis as alternative methods for performance assessments, J. Operational Research Society 44, 1129–1144, 1993) (with one input) and K. F. Lam (In the determination of weight sets to compute cross- efficiency ratios in DEA, J. Operational Research Society 61, 134–143, 2010) (with two inputs). Until now, these data sets have been used by various researchers. The writers did not consider whether or not these data sets were appropriate to the analysis. So, we recommend the CCRCOR model in our paper. The CCRCOR model allows one to add weight restrictions to the CCR model, which are obtained using correlations between input and output variables.