Validation and generalization of DEA and its uses

R. D. Banker, W. W. Cooper, E. Grifell-Tajté, Jesús T. Pastor, Paul W. Wilson, Eduardo Ley, C. A.K. Lovell

    Research output: Contribution to journalArticleResearchpeer-review

    32 Citations (Scopus)


    Rapidly expanding uses of DEA have been accompanied by developments which have enhanced its power and enlarged its utility for additional applications. Developments covered in the present paper include simulation studies comparing DEA with competing forms of statistical regressions. Other studies covered show how these two approaches can be combined in complementary fashion. Another part of this paper deals with Chance Constrained Programming formulations which incorporate probabilistic elements into DEA. Included also are discussions of statistical characterizations with accompanying tests of statistical significance for DEA efficiency evaluations. This paper concludes with uses of DEA in "discovery processes"-processes that need strengthening (and encouragemnt) in contemporary social science and management science research. Suggestions are made for additional research on further developments which extend to uses of DEA to provide new approaches in economics (including econometrics), management and psychology and an Appendix introduces new or recently developed efficiency measures for use in DEA. © 1994 Sociedad de Estadística e Investigación Operativa.
    Original languageEnglish
    Pages (from-to)249-314
    Issue number2
    Publication statusPublished - 1 Dec 1994


    Dive into the research topics of 'Validation and generalization of DEA and its uses'. Together they form a unique fingerprint.

    Cite this