Exploring the Behavior-Driven Crash Risk Prediction Model: The Role of Onboard Navigation Data in Road Safety

Table 6

Logit regression and ANOVA results in datasets with rates of 1 : 1.

Interval (min)	Variable	Estimation	Std	P value	Exp (estimation)
5	A_a_up	2.399	0.85	0.005	11.007
	Constant	−0.203	0.165	0.218	0.816
15	B_a_up	2.48	0.9	0.006	11.943
	B_v_up	−0.055	0.016	0	0.947
	B_n_dn	0.114	0.04	0.005	1.121
	Constant	−0.21	0.14	0.133	0.81
30	A_a_up	2.064	0.88	0.019	7.878
	A_v_up	−0.031	0.012	0.011	0.969
	A_v_dn	−0.029	0.009	0.001	0.972
	B_a_dn	1.549	0.527	0.003	4.708
	Constant	−0.016	0.2	0.935	0.984
45	A_n_dn	−0.076	0.037	0.04	0.927
	A_a_dn	3.136	1.041	0.003	23.016
	A_v_dn	−0.035	0.01	0.001	0.965
	B_a_dn	2.794	0.592	0	16.344
	Constant	−0.633	0.199	0.001	0.531
60	A_a_up	2.757	0.918	0.003	15.753
	A_v_up	−0.045	0.01	0	0.956
	B_a_dn	1.337	0.588	0.023	3.808
	Constant	−0.275	0.203	0.176	0.759