A Quasi-Poisson Approach on Modeling Accident Hazard Index for Urban Road Segments

Table 2

Estimation results for the quasi-Poisson models.

Coefficients for parameters	Fatal-injury (𝜷)		Property-damage-only (𝜹)
	Coefficient	𝑡 stat.	Coefficient	𝑡 stat.
Explanatory variables
Intercept	−2.533	−17.383	−0.713	−3.839
Road segment in rural areas*
Road segment in urban areas	−0.393	−2.844	#	#
Intersection density	0.100	5.922	0.130	11.310
Ownership at state or federal levels*
Ownership at other levels	0.275	1.974	0.441	3.991
Pavement condition rating	#	#	−0.100	−2.141
Percent average daily trucks	−0.083	−4.305	−0.085	−5.071
Dispersion parameter
𝝓	3.096		16.415

# Indicates that the coefficient is statistically insignificant.
*Reference categories.