The Effects of Traffic Composition on Freeway Crash Frequency by Injury Severity: A Bayesian Multivariate Spatial Modeling Approach

Table 3

Parameter estimates in the bivariate CAR model.

Variable	No injury				Injury
	Mean	80%BCI	90%BCI	95%BCI	Mean	80%BCI	90%BCI	95%BCI
Constant	95.24	(90.26, 98.88)	(89.71, 99.26)	(89.42, 99.54)	-19.46	(-24.12, -11.43)	(-24.89, -10.48)	(-25.42, -9.96)
lnDVKT	0.95	(0.70, 1.19)	(0.67, 1.21)	(0.65, 1.22)	0.83	(0.38, 1.17)	(0.29, 1.21)	(0.20, 1.23)
	-0.22	(-0.31, -0.15)	(-0.32, -0.13)	(-0.33, -0.13)	0.20	(0.09, 0.29)	(0.06, 0.30)	(0.04, 0.31)
	-0.62	(-0.74, -0.47)	(-0.77, -0.45)	(-0.79, -.044)	1.75	(1.58, 1.95)	(1.55, 2.00)	(1.52, 2.03)
Curvature	0.09	(0.02, 0.17)	(-0.004, 0.19)	(-0.02, 0.21)	0.10	(-0.04, 0.23)	(-0.08, 0.27)	(-0.11, 0.30)
Gradient	0.04	(-0.13, 0.20)	(-0.18, 0.25)	(-0.22, 0.29)	0.30	(0.02, 0.57)	(-0.05, 0.64)	(-0.11, 0.71)

Bridge and Ramp are excluded, because none of their effects on crash frequencies with the two severity outcomes is statistically significant (less than 80% credibility level).
BCI: Bayesian credible interval.
Boldface indicates statistical significance at the corresponding credibility level.