Evaluating alternate discrete outcome frameworks for modeling crash injury severity

Detecting unforgiving roadside contributors through the severity analysis of ran-off-road crashes

“…This was recognized by Eluru (2013) and Yasmin and Eluru (2013), who recently have compared the performance of both categories of framework models. The main conclusion is that the generalized ordered logit (GOL) model emerges as the true ordered equivalent when compared to the unordered MNL model for examining ordinal discrete variables.…”

“…This advantage applies only to MNL models provided that the dataset under examination satisfies the IIA property, which is the case for our data set. Hence, the mixed logit, by relaxing the IIA property, is unlikely to yield unbiased estimates in the event of underreporting (Yasmin and Eluru, 2013).…”

Factors associated with urban non-fatal road-accident severity

“…A study investigated associations between accident severity and the age of motorcycle drivers, which were grouped into 'under 25' and 'over 55' years using the age group of '25-55' as reference (Patil, Geedipally, & Lord, 2012), whereas another study classified drivers into five age groups (<18, 18-30, 31-45-46-60 and 60+ year of age) (Theofilatos et al, 2012). Regarding the inconsistency of findings, two studies reported that male drivers had a lower injury propensity compared to female drivers in their study of 60 jurisdictions across the US and the city of Coimbra, Portugal, respectively (Garrido, Bastos, de Almeida, & Elvas, 2014;Yasmin & Eluru, 2013). On the other hand, Zhang, Lindsay, Clarke, Robbins, and Mao (2000) reported that elderly (65+ years of age) male drivers in Ontario, Canada, were more likely to be involved in a major or fatal accident than females.…”

Analysis of Factors Influencing the Vehicle Damage Level in Fatal Truck-Related Accidents and Differences in Rural and Urban Areas

“…The response variable vehicle damage level is inherently ordinal discrete, so a categorical response model is used to identify the factors affecting it. Commonly, statistical formulation to model vehicle damage level is the ordered response formulation, especially ordered logit model or ordered probit model [4,5,6,7,8,9,10,11,12,13]. The main difference between the two ordered models is the assumption of distribution of the error term.…”