Page 54 - tmp_Neat

According to the VicRoads classification, crash injury severity is classified into four
categories: fatal injury (killed or died within 30 days), serious injury (sent to hospital,

possibly admitted), other injury (typically requires medical treatment) and non-injury
(property damage only (PDO). From 2006 to 2016, 20,957 people were involved in 8,486

crashes involving heavy vehicles in Victoria. Of the 20,957 people involved, 480 (2.3%)
were fatally injured, 3,697 (17.6%) suffered severe injury, 6,298 (30.1%) had minor injury

and 10,482 (50.0%) experienced no injury.

4.2.2 Multinomial Logit Model

The summary of existing studies on multinomial logit model is presented in Section 2.2.1.3

of Chapter 2. The dependent variable in this study is the road user injury severity in crashes
involving heavy vehicles, which is classified into four categories in this research: fatal, severe

injury, minor injury and non-injury. Considering the categorical nature of the dependent

variable, the use of the multinomial logit regression model is appropriate. This model has
been widely applied by previous researchers in road safety for modelling dependent variables

with more than two outcomes (Khorashadi et al., 2005: Tay et al., 2011; Eluru, 2013; Ye and
Lord, 2014; Wu et al. 2016).

Let ( ) be the probability of collision n ending in injury severity category i, then

( ) ( ) (4.1)

where,
is a vector of measurable characteristics

is a vector of coefficients to be estimated

is an error term accounting for unobserved effects influencing injury severity.

If the error terms are assumed to be type 1 extreme value distribution (McFadden, 1981), then

( )
( ) ∑ ( ) (4.2)

43