Identifying the Smoking and Smokeless Tobacco-Related Predictors on Frequencies of Heavy Vehicle Traffic Accidents in Bangladesh: Linear and Binary Logistic Regression-Based Approach
Table 4
Coefficients and the model summary of the regression, fits, and diagnostics for unusual observations and prediction for response.
Term
SE coef
T value
value
VIF
Coefficients
Constant
0.144
10.10
≤0.001
Foursmokeormore
YES
1.62
3.02
0.003
1.01
SmokeDurDrive
YES
0.225
2.89
0.004
1.01
Model summary
R-sq
R-sq (adj)
R-sq (pred)
4.33%
3.87%
≤0.001%
Fits and diagnostics for unusual observations
Obs
AccidentFreq
Fit
Residuals
Std residuals
Type of residuals
34
9.000
1.458
7.542
3.32
R
36
12.000
7.000
5.000
3.10
X
44
9.000
2.110
6.890
3.03
R
54
7.000
2.110
4.890
2.15
R
55
9.000
2.110
6.890
3.03
R
60
9.000
1.458
7.542
3.32
R
68
12.000
1.458
10.542
4.64
R
102
9.000
2.110
6.890
3.03
R
104
9.000
1.458
7.542
3.32
R
121
7.000
2.110
4.890
2.15
R
124
12.000
2.110
9.890
4.35
R
133
10.000
1.458
8.542
3.76
R
140
7.000
1.458
5.542
2.44
R
150
2.000
7.000
−5.000
−3.10
X
216
12.000
2.110
9.890
4.35
R
217
8.000
1.458
6.542
2.88
R
218
10.000
1.458
8.542
3.76
R
242
8.000
2.110
5.890
2.59
R
295
9.000
1.458
7.542
3.32
R
308
7.000
2.110
4.890
2.15
R
379
12.000
2.110
9.890
4.35
R
380
9.000
1.458
7.542
3.32
R
406
9.000
2.110
6.890
3.03
R
408
7.000
2.110
4.890
2.15
R
411
10.000
2.110
7.890
3.47
R
Prediction for AccidentFreq
Variable
Setting
Fit
SE fit
95% CI
95% PI
Foursmokeormore
NO
1.4578
0.144363
(1.17407, 1.74159)
(−3.02885, 5.94452)
XX
SmokeDurDrive
NO
Foursmokeormore
YES
6.3480
1.62650
(3.15092, 9.54509)
(0.84608, 11.8499)
XX
SmokeDurDrive
NO
Foursmokeormore
YES
7
1.61080
(3.83379, 10.1662)
(1.51596, 12.4840)
XX
SmokeDurDrive
YES
Foursmokeormore
NO
2.1098
0.173194
(1.76939, 2.45026)
(−2.38080, 6.60045)
XX
SmokeDurDrive
YES
∗∗R = large std residuals, X = unusual std residuals, and XX denotes an extremely unusual point relative to predictor levels used to fit the model.
