Application of the Bayesian Model Averaging in Analyzing Freeway Traffic Incident Clearance Time for Emergency Management

Table 2

Selected models with the highest posterior probabilities using BMA.

Variable		EV^a	SD^b	Model1	Model2	Model3	Model4	Model5
Response time	100.0	−0.092	0.004	−0.091	−0.094	−0.091	−0.092	−0.092
Traffic control	100.0	−0.527	0.062	−0.536	−0.509	−0.533	−0.502	−0.533
Collision	100.0	−0.737	0.211	−0.866	−0.780	−0.877	−0.380	−0.868
Single lane blocked	92.2	−0.182	0.078	−0.173	−0.217	−0.176	−0.216	−0.182
Multiple lane blocked	100.0	−0.619	0.111	−0.608	−0.670	−0.586	−0.639	−0.623
Total closure	100.0	−1.491	0.357	−1.501	−1.531	−1.536	−1.444	−1.523
Injury involved	100.0	−0.485	0.123	−0.486	−0.484	−0.475	−0.500	−0.484
Fire involved	92.6	−1.061	0.455	−1.248	−1.177	−1.143	−8.969	−1.257
Summer	100.0	0.170	0.045	0.176	0.171	0.173	0.163	0.173
nVar^c				11	12	10	12	12
BIC				−1488	−1486	−1486	−1486	−1485
PMP				0.228	0.117	0.106	0.086	0.060

^aEV is posterior means. ^bSD means the standard deviation. ^cnVar refers to the number of variables in the model.