Application of the Empirical Bayes Method with the Finite Mixture Model for Identifying Accident-Prone Spots

Table 4

Parameter estimates for the FMNB-2 and GFMNB-2 models.

Method	Component	Statistic	Ln()
Period 1
FMNB-2	1	Estimate	−6.275	0.991	−0.056	−0.013	−0.188	0.626
		SE	0.628	0.067	0.025	0.005	0.024	0.091
	2	Estimate	−10.026	1.248	−0.035	−0.014	0.216	0.450
		SE	0.660	0.072	0.025	0.005	0.016	0.115
GFMNB-2	1	Estimate	−6.045	0.830	−0.044	−0.011	0.079	0.482
		SE	0.628	0.066	0.026	0.005	0.016	0.107
	2	Estimate	−3.906	0.669	−0.027	−0.024	0.080	0.894
		SE	0.899	0.103	0.027	0.005	0.028	0.087
	Estimate		64.211	−199.09	0.019	−6.908	1.499	−18.618
	Period 2
FMNB-2	1	Estimate	−7.424	1.113	−0.049	−0.015	0.027	0.833
		SE	0.685	0.071	0.028	0.006	0.020	0.075
	2	Estimate	−8.085	1.086	−0.068	−0.009	0.029	0.247
		SE	0.473	0.051	0.018	0.004	0.014	0.125
GFMNB-2	1	Estimate	−3.138	0.715	−0.089	−0.036	0.067	0.708
		SE	0.731	0.077	0.026	0.005	0.021	0.088
	2	Estimate	−7.111	1.004	−0.082	−0.013	0.041	0.344
		SE	0.487	0.051	0.020	0.004	0.014	0.091
	Estimate		2.282	4.8630	−0.0003	−0.091	−0.066	0.187

Not significant at 5% significance level; SE: standard error.