Generally taken to mean an odd unit with an input-output mix that significantly differs from the rest of the units. This can be due to a measurement error, or as a result of the outlier having different operating practices. Outliers that are efficient can introduce bias into the analysis. See Epstein and Henderson (1989). [Epstein M K and Henderson J C (1989) ‘Data envelopment analysis for managerial control and diagnosis’, Decision Sciences, 20, pp 90-119.]

Leave a Reply

goin up