Data-Driven Urban Traffic Accident Analysis and Prediction Using Logit and Machine Learning-Based Pattern Recognition Models

Table 4

Variables of the severity of accident logit model.

Variables	β	Standard deviation	Wald statistic	Significance	Odds ratio
6 : 00 to 12 : 00	−3.345	1.326	6.364	0.012	0.035
12 : 00 to 18 : 00	2.939	1.286	5.201	0.023	18.897
18 : 00 to 24 : 00	7.373	2.488	8.783	0.003	1592.404
Autumn	−1.918	0.725	7.010	0.008	0.147
Sunny weather	−2.446	0.806	9.205	0.002	0.087
Rainy weather	2.751	1.157	5.656	0.017	15.658
Lack of attention	1.119	0.644	3.020	0.082	3.062
Female driver	−2.063	0.783	6.946	0.008	0.127
Head-on collision	2.204	0.771	8.176	0.004	9.061
Day time	−4.343	1.422	9.323	0.002	0.013
KIA Pride vehicle	3.795	1.287	8.696	0.003	44.478
Constant	6.734	2.914	5.338	0.021	840.503