This page is published in English.
DEA
DEA-RAM - Range-Adjusted Measure of Inefficiency
Non-parametric efficiency - Non-radial, weighted slack-based (VRS)
Cooper, W. W., Park, K. S., Pastor, J. T.1999doi:10.1023/A:1007701304281 ↗
Overview
RAM is a non-radial DEA model using range-adjusted weights for input/output slacks. Unlike CCR/BCC (radial), RAM accounts for all slacks simultaneously. ρ=1 means fully efficient; ρ<1 quantifies normalised distance to frontier. Requires VRS (Σλ=1). Insensitive to units of measurement.
- Output
- Efficiency, higher is better
- Data
- Crisp
- Size
- 3+ alternatives, p+s ≤ m/3 criteria works best
- Used for
- Education performance, healthcare benchmarking, public sector efficiency, returns to scale identification
Look elsewhere when
- •CRS assumption holds - use CCR
- •Negative data present
- •All values in a column identical (zero-range column)
Assumptions to verify
- All inputs and outputs non-negative
- No constant column (zero range would produce undefined weight)
- VRS assumption is appropriate (Σλ=1 enforced)
Edge cases and pitfalls
Constant input/output column causes zero range weight (division by zero in weight computation). Remove or merge such variables.
How to cite
Cooper, W. W.; Park, K. S.; Pastor, J. T. (1999). RAM: A Range Adjusted Measure of Inefficiency for Use with Additive Models, and Relations to Other Models and Measures in DEA. Journal of Productivity Analysis. https://doi.org/10.1023/A:1007701304281
System ID, as it appears in reports and the API
DEA-RAM