Title: Cannot be more efficient than the Pareto efficiency?
1Cannot be more efficient than the Pareto
efficiency?
- Lifen Wu
- Centre for Efficiency and Productivity Analysis
- The University of Queensland
- Australia
- Philadelphia, July 10 12, 2009
2DEA calculations are Pareto optimal
3DEA efficient conditions are those of Pareto
efficiency
4Technical and scale efficiencies
5VRS model with Mixed-Orientation
6Projection to be scale efficient
7Tim Coellis example of 3 inputs and 1 output
8Maximizing sum of slacks derived from strict
positivityvia non-Archimedean constant (BCC
models)
9Maximizing sum of slacks derived from strict
positivityvia non-Archimedean constant (CCR
models)
10Origin of Strict Positivity (1979)
11Expression (2) in Measuring the efficiency of
decision making units (1978)
12DMU0 may not satisfy the 2nd condition
13Strict positivity results in redundant constraint
14 or makes original CCR model oriented
15Charnes and Cooper aware of this
16Proposed condition 2 of Pareto-efficiency
- Zero is the only possible solution for all the
slack variables in envelopment form
17Conclusion
- DEA calculations can be more optimal (efficient)
than the Pareto optimality (efficiency)
18Pareto efficiency and inequality