Niels Chr. Petersen

The Use of Data Envelopment Analysis with Probabilistic Assurance Regions for Measuring Hospital Efficiency

This paper uses Data Envelopment Analysis (DEA) for an estimation of the cost efficiency of 70 Danish hospitals. The analysis relates to a cost function based on 483 outputs in combination with a set of probabilistic assurance regions defined by the cost distributions for each output. It is demonstrated that the probabilistic assurance region approach allows for (i) a frontier estimation in the full output space, i.e., no fixed aggregation is required, and (ii) a controlling of the variation in heterogeneity of the output clusters, in casu Diagnosis Related Groups. The likelihood of the estimated efficiency score for a given hospital can be measured based on the sensitivity of the score w.r.t. the probability levels used in the specification of confidence intervals for the probabilistic assurance regions.

Facet Analysis in Data Envelopment Analysis

Data Envelopment Analysis (DEA) employs mathematical programming to measure the relative efficiency of Decision Making Units (DMUs). One of the topics of this chapter is concerned with development of indicators to determine whether or not the specification of the input and output space is supported by data in the sense that the variation in data is sufficient for estimation of a frontier of the same dimension as the input output space. Insufficient variation in data implies that some inputs/outputs can be substituted along the efficient frontier but only in fixed proportions. Data thus locally support variation in a subspace of a lower dimension rather than in the input output space of full dimension. The proposed indicators are related to the existence of so-called Full Dimensional Efficient Facets (FDEFs). To characterize the facet structure of the CCR- or the BCC-estimators, (Charnes et al. Eur J Oper Res 2:429–444, 1978; Banker et al. Manage Sci 30(9):1078–1092, 1984) of the efficient frontier we derive a dual representation of the technologies. This dual representation is derived from polar cones. Relying on the characterization of efficient faces and facets in Steuer (Multiple criteria optimization. Theory, computation and application, 1986), we use the dual representation to define the FDEFs. We provide small examples where no FDEFs exist, both for the CCR- and the BCC estimator. Thrall (Ann Oper Res 66:109–138, 1996) introduces a distinction between interior and exterior facets. In this chapter we discuss the relationship between this classification of facets and the distinction in Olesen and Petersen (Manage Sci 42:205–219, 1996) between non-full dimensional and full dimensional efficient facets. Procedures for identification of all interior and exterior facets are discussed and a specific small example using Qhull to generate all facets is presented. In Appendix B we present the details of the input to and the output from Qhull. It is shown that the existence of well-defined marginal rates of substitution along the estimated strongly efficient frontier segments requires the existence of FDEFs. A test for the existence of FDEFs is developed, and a technology called EXFA that relies only on FDEFs and the extension of these facets is proposed, both in the context of the CCR-model and the BCC-model. This technology is related to the Cone-Ratio DEA. The EXFA technology is used to define the EXFA efficiency index providing a lower bound on the efficiency rating of the DMU under evaluation. An upper bound on the efficiency rating is provided by a technology defined as the (non-convex) union of the input output sets generated from FDEFs only. Finally, we review recent uses of efficient faces and facets in the literature.