Estimation of Cost Efficiency in Non-parametric Frontier Models

Abstract

The paper proposes a bootstrap methodology for robust estimation of cost efficiency in data envelopment analysis. Our algorithm re-samples "naive" input-oriented efficiency scores, rescales original inputs to bring them to the frontier, and then re-estimates cost efficiency scores for the rescaled inputs. We consider the cases with absence and presence of environmental variables. Simulation analyses with multi-input multi-output production function demonstrate consistency of the new algorithm in terms of the coverage of the confidence intervals for true cost efficiency. Finally, we offer real data estimates for Japanese banking industry. Using the nationwide sample of Japanese banks in 2009, we show that the bias of cost efficiency scores may be linked to the bank charter and the presence of the environmental variables in the model. A package `rDEA', developed in the R language, is available from the GitHub and CRAN repository.