Binary Logistic Regression Modeling of Idle CO Emissions in Order to Estimate Predictors Influences in Old Vehicle Park
Table 4
Logistic estimates for CO emission test failure.
Model
Predictors
Wald statistics (df)
SRB
VA
0.077
0.014
31.08 ()a
1.080
EP()
−0.766
0.173
19.580 ()a
0.465
EP()
−1.14
0.29
15.453 ()b
0.319
SFS()
1.938
0.585
10.961 ()a
6.944
K()
2.943
1.042
7.969 ()b
18.969
EU
OR
0.000017
0.000001
217.803 ()a
1.000017
EP()
−1.636
0.157
108.538 ()a
0.195
EP()
−2.319
0.24
93.342 ()a
0.098
SFS()
−7.106
1.05
45.829 ()a
0.00082
K()
4.502
1.031
19.067 ()a
90.181
USA
OR
0.000045
0.000006
56.939 ()a
1.000045
VA
0.597
0.078
59.413 ()a
1.822
EP()
−6.436
0.98
43.167 ()a
0.001603
EP()
−8.525
1.166
53.485 ()a
0.000199
SFS()
9.5
1.366
48.339 ()a
1,3360.72
K()
6.41
1.246
26.483 ()a
608.141
Notes: value ≤ 0.01; value ≤ 0.05; df: degree of freedom; EP(): first subcategory in category of engine power (41–70 kW); EP(): second subcategory in the category of engine power (>71 kW); SFS(): first subcategory in category of the system of fuel supply (vehicles with carburetor); K(): first subcategory in the category of the catalytic converter (CAT) (vehicles without CAT).
