The nonparaeff package's function fdh(), when used with an output-orientation, produces an efficiency score that puzzles me. I am using nonparaeff in R version 2.8.1 running under Linux. In this context, I ran the following code: library(nonparaeff) input= c(6, 10, 16, 14, 15) output=c(8, 20, 26, 4, 24) tab.dat = data.frame(y=output, x=input) fdhoo = fdh(tab.dat, noutput=1, orientation=2) fdhoo That code produced the following output: eff 1 1 2 1 3 1 4 1 5 1 The efficiency score for DMU 4 surprises me. This DMU, using input quantity 14, should be able to produce at least as much as DMU 2, which uses input level 10. Yet DMU 4 produces 4 output units, compared to 20 for DMU 2. I would expect DMU 4 to be able to raise output by a factor of 5. If so, the reported eff is puzzling. Can anyone tell me how to get nonparaeff to produce correct efficiency scores in output orientation? I'll be very grateful for your suggestions. Best regards, John -- John P. Burkett Department of Economics University of Rhode Island Kingston, RI 02881-0808 USA phone (401) 874-9195