Data Envelopment Analysis (DEA) package for R, estimating robust DEA scores without and with environmental variables and doing returns-to-scale tests.
The package can be installed from the official R CRAN by
Under Linux make sure you have GLPK installed before, e.g., in Ubuntu you need
Or if you need to compile GLPK set following environmental variables before installing rDEA
Using devtools our package can be installed by (make sure you have libglpk-dev installed)
In Linux and Mac another option is to use command line (make sure you have slam, truncreg and truncnorm R packages installed)
Robust DEA (Simar and Wilson, 1998) with input model with included Japan hospital data
library(rDEA)
## loading Japan hospital data
data("hospitals", package="rDEA")
## choosing inputs and outputs for analysis
firms = 1:50
Y = hospitals[firms, c('inpatients', 'outpatients')]
X = hospitals[firms, c('labor', 'capital')]
## Robust DEA with 1000 bootstrap iterations and variable returns-to-scale
di = dea.robust(X=X, Y=Y, model="input", RTS="variable", B=1000)
## robust estimates of technical efficiency for each hospital
di$theta_hat_hat
After installing rDEA package you can run included tests by