Abstract
Data envelopment analysis (DEA) has become a popular approach to nonparametric efficiency measurement. The statistical inference using bootstrap methods is readily available for the radial DEA estimator; however it is missing for the Russell measure, the nonradial DEA estimator. We propose a bootstrap based procedure for making statistical inference about the individual Russell measures of technical efficiency. We perform simulations to examine finite sample properties of the proposed estimator. Finally, we present an empirical study using proposed bootstrap procedure.
Acknowledgements
The authors wish to thank two anonymous Referees and the Editor for their valuable comments. The usual disclaimers apply. The authors acknowledge the Lebesgue Center of Mathematics who supported the work of Second Author according to the program PIA-ANR-11-LABX-0020-01 during his postdoc at Agrocampus Ouest, when a substantial part of this project was carried out.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Though we do not directly set the correlation between λ1 and λ2, our experiments with different constellations gave us a sense that setting the correlation between λ10 and λ20 directly influenced the correlation between λ1 and λ2.