Predicting Wet-Road Crashes Using the Finite-Mixture Zero-Truncated Negative Binomial Model

Table 3

Estimating results of the FMZTNB-2 model.

Variable	Estimate	Std. err.	value	Estimate	Std. err.	value
	Component 1			Component 2
Intercept	−4.047	2.819	0.151	−4.485	1.190	<0.001
Log (ADT)	0.718	0.018	<0.001	0.993	0.195	<0.001
Lane width	−0.080	0.010	<0.001	0.203	0.012	<0.001
Ave. Out. SHD	−0.094	0.022	<0.001	−0.183	0.014	<0.001
Ave. In. SHD	−0.241	0.177	0.174	−0.879	0.166	<0.001
Median width	−0.416	0.010	<0.001	−0.494	0.251	0.049
Disp. par.	1.499	0.142	<0.001	4.619	1.856	0.013
Weight	0.712	0.082	<0.001	0.288	–	–
AIC	1020.54	–	–	–	–	–
BIC	1088.32	–	–	–	–	–
MAE	0.22	–	–	–	–	–
RMSE	2.18	–	–	–	–	–

Note. Std. err. = standard error. Disp. Par. = dispersion parameter. Bold and italic values indicate statistically significant at level of 90.0 percent or higher. — means not applicable.