dea: Data envelopment analysis

In tkmckenzie/snfa: Smooth Non-Parametric Frontier Analysis

Description

Basic data envelopment analysis to replace rDEA::dea due to difficulties installing rDEA on travis-ci

Usage

dea(XREF, YREF, X, Y, model = "output", RTS = "variable")

Arguments

XREF	Matrix of inputs for observations used for constructing the frontier
YREF	Matrix of outputs for observations used for constructing the frontier
X	Matrix of inputs for observations, for which DEA scores are estimated
Y	Matrix of outputs for observations, for which DEA scores are estimated
model	Orientation of the model; must be "input" or "output"
RTS	Returns-to-scale for the model; must be "constant", "non-increasing" or "variable"

Details

This function estimates efficiency using data envelopment analysis. The linear program is constructed as in Fare et al. (1985) and optimized using lpSolve::lp. The function was built to swap with rDEA::dea because installation of rDEA presented problems for travis-ci, due to difficulties installing glpk. Should installation of rDEA on travis-ci become possible, this function will be removed.

Value

Returns a list with the following components

thetaOpt	A vector of efficiency estimates, in [0, 1] interval
lambda	A matrix of constraint coefficients
lambda_sum	A vector of sum of lambdas; lambda_sum = 1 for variable RTS, lambda_sum <= for non-increasing RTS
model	Orientation of the model
RTS	Returns-to-scale for the model

References

\insertRef

FareDEAsnfa

Examples

data(univariate)

dea.fit <- dea(univariate$x, univariate$y,
               univariate$x, univariate$y,
               model = "output",
               RTS = "variable")
univariate$frontier <- univariate$y / dea.fit$thetaOpt

# Plot technical/allocative efficiency over time
library(ggplot2)

ggplot(univariate, aes(x, y)) +
  geom_point() +
  geom_line(aes(y = frontier), color = "red")

tkmckenzie/snfa documentation built on June 11, 2020, 4:34 a.m.