Method for Estimating Confidence Intervals for DEA Efficiency Models Using Regression Models

  • Filipe Giovani Bonin BisoffiEmail author
  • Graziella Cardoso Bonadia
  • Victor Henrique Duarte de Oliveira
  • Sérgio Ricardo Barbosa
Part of the Telecommunications and Information Technology book series (TIT)


Benchmarking methods, such as DEA (Data Envelopment Analysis), are used to compare a set of entities regarding their efficiency in a given process. The structure of the DEA method does not take random disturbances into consideration when estimating the efficiency of each entity. In most scenarios, this characteristic does not reflect the reality of the problem, since practically the entire process is subject to external disturbances. Using regression methods, it is possible to generate confidence intervals for DEA-estimated efficiency, considering the model’s inputs and outputs as independent variables. With this, the conclusions and subsequent actions based on the returned results are more robust, and begin to contemplate, in a certain manner, random disturbances suffered by the companies.


DEA Confidence interval Nonparametric regression analysis 


  1. 1.
    Molinero CM, Woracker D (1996) Data envelopment analysis. Or Insight 9(4):22–28Google Scholar
  2. 2.
    Lorenzett JR, Lopes ALM, De Lima MVA (2010) Aplicação de método de pesquisa operacional (DEA) na avaliação de desempenho de unidades produtivas para área de educação profissional. Revista Eletrônica de Estratégia & Negócios 3(1):168–190Google Scholar
  3. 3.
    Kumbhakar SC, Lovell CAK (2003) Stochastic frontier analysis. Cambridge University PressGoogle Scholar
  4. 4.
    Bogetoft P, Otto L (2010) Benchmarking with DEA, SFA, and R. SpringerGoogle Scholar
  5. 5.
    Souza G, Souza M, Gomes E (2011) Computing confidence intervals for output-oriented DEA models: an application to agricultural research in Brazil. J Oper Res Soc 62(10):1844–1850Google Scholar
  6. 6.
    Cribari-Neto F, Zeileis A (2009) Beta regression in RGoogle Scholar
  7. 7.
    Grün B, Kosmidis I, Zeileis A (2011) Extended beta regression in R: shaken, stirred, mixed, and partitioned. Working papers in economics and statisticsGoogle Scholar
  8. 8.
    Amemiya T (1984) Tobit models: a survey. J Econom 24(1):3–61Google Scholar
  9. 9.
    Fox J (2002) An R and S-plus companion to applied regression. SageGoogle Scholar
  10. 10.
    Banker RD, Charnes A, Cooper WW (1984) Some models for estimating technical and scale inefficiencies in data envelopment analysis. Manag Sci 30(9):1078–1092Google Scholar
  11. 11.
    Nadaraya EA (1964) On estimating regression. Theor Probab Appl 9(1):141–142Google Scholar
  12. 12.
    Epanechnikov VA (1969) Non-parametric estimation of a multivariate probability density. Theor Probab Appl 14(1):153–158Google Scholar
  13. 13.
    Comaniciu D, Meer P (1999) Mean shift analysis and applications. In: The proceedings of the seventh IEEE international conference on computer vision. IEEE, pp 1197–1203Google Scholar
  14. 14.
    Wand MP, Jones MC (1994) Kernel smoothing. CRC PressGoogle Scholar
  15. 15.
    Turlach BA et al (1993) Bandwidth selection in kernel density estimation: a review. Université catholique de LouvainGoogle Scholar
  16. 16.
    Li Q, Racine J (2004) Cross-validated local linear nonparametric regression. Stat Sin 14(2):485–512Google Scholar
  17. 17.
    Hayfield T, Racine JS (2008) Nonparametric econometrics: the np package. J Stat Softw 27(5).
  18. 18.
    Efron B (1979) Bootstrap methods: another look at the jackknife. In: The annals of statistics, pp 1–26Google Scholar
  19. 19.
    Morettin PA, Toloi C (2006) Análise de séries temporais. BlucherGoogle Scholar
  20. 20.
    Lothgren M (1998) How to bootstrap DEA estimators: a Monte Carlo comparison. WP in Economics and Finance, no 223Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Filipe Giovani Bonin Bisoffi
    • 1
    Email author
  • Graziella Cardoso Bonadia
    • 1
  • Victor Henrique Duarte de Oliveira
    • 2
  • Sérgio Ricardo Barbosa
    • 2
  1. 1.Decision Management Support & Applications DepartmentCPqD - Telecommunications Research and Development CenterCampinasBrazil
  2. 2.CEMIG Distribuição S.A.Belo HorizonteBrazil

Personalised recommendations