Ordinal principal component analysis for a common ranking of stochastic frontiers

ABSTRACT

The Stochastic Frontier Analysis (SFA) is a model to evaluate the Technical Efficiency (TE) for Production Units (PU). When SFA is applied on different output variables with same input, the analysis estimates different TEs for the PU. We refer to these TEs as the Multiple Technical Efficiency (MTE) of the PU. In this work, we present a method to unify the MTE in one ranking, in order to compute a synthetic index of the TE based on a parametric model. Our approach transforms the measures of efficiency into values on an ordinal scale. Then, using the Ordinal Principal Component Analysis and a genetic algorithm, we merge the multiple rankings.

KEYWORDS:

• Technical efficiency
• multiple technical efficiency
• ordinal scale
• production unit
• ranking

Acknowledgements

Thanks to the referees for the suggestions that have certainly improved the proposed work.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Sergio Scippacercola http://orcid.org/0000-0003-4416-5838

Enrica Sepe http://orcid.org/0000-0002-1970-8258

Additional information

Funding

This work has been partially supported by the Nidge University under Grant number 14651.