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

Table 5

Summary statistics of each component for Periods 1 and 2.

Method	Component (sample)	Statistic	Crashes		LW	SW	CD
Period 1 (years 1997 and 1998)
Mean-based classification method	Component 1 (1007)	Mean	0.68	5461.00	12.63	10.62	1.45	0.40
		SD	0.77	3376.75	1.60	7.95	2.53	0.40
	Component 2 (492)	Mean	7.51	8296.00	12.46	8.61	1.39	0.87
		SD	6.15	4012.35	1.55	7.98	1.92	0.93
FMNB-2	Component 1 (545)	Mean	3.91	6274	12.51	9.99	1.69	0.42
		SD	4.76	3573.48	1.52	7.99	2.85	0.45
	Component 2 (954)	Mean	2.37	6458	12.61	9.95	1.28	0.63
		SD	4.75	3977	1.63	8.03	2	0.75
GFMNB-2	Component 1 (738)	Mean	3	8191	12.58	11.22	0.7	0.29
		SD	4.45	3867.52	1.62	8.31	1.54	0.17
	Component 2 (761)	Mean	2.85	4646	12.57	8.74	2.14	0.81
		SD	5.13	2878.96	1.56	7.53	2.75	0.85
Period 2 (years 1999 to 2001)
Mean-based classification method	Component 1 (421)	Mean	12.92	9420.99	12.46	8.51	1.39	0.96
		SD	10.79	4374.44	1.59	7.97	1.82	0.98
	Component 2 (1078)	Mean	1.33	5723.28	12.62	10.53	1.44	0.39
		SD	1.31	3556.24	1.58	7.97	2.53	0.39
FMNB-2	Component 1 (289)	Mean	10.29	7601.30	12.68	9.89	1.44	0.38
		SD	10.72	4495.77	1.60	7.97	2.46	0.34
	Component 2 (1210)	Mean	3.22	6561.29	12.55	9.98	1.43	0.60
		SD	6.21	4039.09	1.58	8.03	2.32	0.72
GFMNB-2	Component 1 (452)	Mean	6.27	9145.69	12.95	12.98	0.85	0.26
		SD	8.60	4457.79	1.74	8.12	1.90	0.15
	Component 2 (1047)	Mean	3.85	5732.65	12.41	8.66	1.68	0.68
		SD	7.33	3546.66	1.49	7.61	2.48	0.76