A Distribution-Free Approach to Stochastic Efficiency Measurement with Inclusion of Expert Knowledge

Table 5

DEA and CCP efficiency evaluation for simulation 3: if the data contains small nonsystematic errors, the DEA model outperforms the CCP. CCP works well under conditions where inefficiency has not been partially offset by noise.

	DEA; 𝜃	∑^𝑞𝑅=1𝜆𝑟,𝑅	CCP (U); 𝜃^𝑈CCP	CCP (L); 𝜃^𝐿CCP	Average; 𝜃CCP	CCP; 𝜃CCP	∑^𝑞𝑅=1𝜆^𝑈𝑟,𝑅	∑^𝑞𝑅=1𝜆^𝐿𝑟,𝑅	∑^𝑞𝑅=1𝜆𝑟,𝑅
DMU1	1	1	0.794	1.566	1.18	1	1.136 (4)	1.283 (2)	1.20945
DMU2	1	1.901	0.731	1.603	1.167	1	3.148 (11)	1.305 (4)	2.22655
DMU3	0.845	0	0.659	1.003	0.831	0.831	0	0	0
DMU4	0.898	0	0.67	1.079	0.874	0.874	0	0	0
DMU5	1	2.986	0.728	1.235	0.982	0.982	0 (0)	2.137 (7)	1.0685
DMU6	0.779	0	0.571	0.954	0.762	0.762	0	0	0
DMU7	1	2.704	0.725	1.24	0.982	1	5.681 (10)	2.598 (7)	4.13975
DMU8	0.877	0	0.705	1.028	0.867	0.867	0	0	0
DMU9	1	1	0.791	2.408	1.599	0.896	0 (0)	1.963 (10)	0.98141
DMU10	0.779	0	0.664	0.88	0.772	0.772	0	0	0
DMU11	1	1	0.799	1.298	1.048	0.899	0	0.6928	0.3464
DMU12	0.814	0	0.674	0.926	0.8	0.8	0	0	0
DMU13	1	2.409	0.893	1.161	1.027	0.947	2.634 (8)	0.451 (3)	1.54245
DMU14	1	1	0.936	29.92	15.43	0.968	0.585 (2)	3.528 (6)	2.05655
DMU15	1	1	0.926	2.77	1.848	1	1.816 (2)	1.041 (2)	1.42865