Environmental Model

In this example we compute the Aparicio, Pastor and Zofio (2023) environmental model for inputs X, good outputs Y, and bad outputs B, using directions Gx, Gy, and Gb. If not specified, default directions are Gx = :Zeros, Gy = :Observed, Gb = :Observed.

using DataEnvelopmentAnalysis

X = ones(5, 1);
Y = [7; 5; 1; 3; 4];
B = [2; 5; 3; 3; 2];

deaenv1 = deaenv(X, Y, B)

Environmental DEA Model 
DMUs = 5; Inputs = 1; Outputs = 1; Bad Outputs = 1
Returns to Scale = CRS
Gx = Zeros; Gy = Observed ; Gb = Observed
──────────────────────────────────────────────
   efficiency   slackX1       slackY1  slackB1
──────────────────────────────────────────────
1    0.0       0.0        0.0              0.0
2    0.4       0.0       -6.21725e-16      1.0
3    0.826087  0.73913   -1.38778e-15      0.0
4    0.555556  0.333333  -1.11022e-15      0.0
5    0.272727  0.272727   0.0              0.0
──────────────────────────────────────────────

Estimated efficiency scores are returned with the efficiency function:

efficiency(deaenv1)

5-element Vector{Float64}:
 0.0
 0.4000000000000002
 0.8260869565217392
 0.5555555555555556
 0.27272727272727276

The optimal peers, $λ$, are returned with the peers function and are returned as a DEAPeers object:

peers(deaenv1)

DEA Peers
1: 1 ( 1.0 ) 
2: 1 ( 1.0000000000000002 ) 
3: 1 ( 0.2608695652173913 ) 
4: 1 ( 0.6666666666666667 ) 
5: 1 ( 0.7272727272727273 ) 

deaenv Function Documentation

deaenv(X, Y, B; Gx, Gy, Gb)