Evaluating temporal variability of exogenous variable impacts over 25 years: An application of scaled generalized ordered logit model for driver injury severity

Marcoux, Robert; Yasmin, Shamsunnahar; Eluru, Naveen; Rahman, Moshiur

doi:10.1016/j.amar.2018.09.001

Cited by 40 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future research could focus on the investigation of spatial or temporal unobserved variations that can potentially be present in the analysis of the SHRP2 NDS data. In this context, spatial effects models (Aguero-Valverde et al, 2016;Tischer et al, 2019) or scaled logit/generalized logit models (Swait and Louviere, 1993;Marcoux et al 2018) could be leveraged. Such models may have the potential to identify specific temporal or spatial effects, which are likely to be captured as pure unobserved heterogeneity, even with the use of advanced random parameter models.…”

Section: Discussionmentioning

confidence: 99%

Do high visibility crosswalks improve pedestrian safety? A correlated grouped random parameters approach using naturalistic driving study data

Pantangi

Ahmed

Fountas

et al. 2021

Analytic Methods in Accident Research

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Do high visibility crosswalks improve pedestrian safety? A correlated grouped random parameters approach using naturalistic driving study data

Pantangi

Ahmed

Fountas

et al. 2021

Analytic Methods in Accident Research

View full text Add to dashboard Cite

“…However, some variables affecting the bicycle crash levels at various intersections may be different. Following the recent studies such as Marcoux et al [33], the generalized ordered logit model (GOL), which can relax the PO assumption for all variables, is selected in this study. Actually, in this study, we are not convinced whether we need to relax the PO constraint for all or some specific variables.…”

Section: Methodologiesmentioning

confidence: 99%

Determinants of Bicyclist Injury Severity Resulting from Crashes at Roundabouts, Crossroads, and T-Junctions

Shen

Wang

Zheng

et al. 2020

Journal of Advanced Transportation

View full text Add to dashboard Cite

This study explores the contributing factors that influence bicyclist injury severity at three types of intersection: roundabouts, crossroads, and T-junctions. Using bicycle-involved crash data in the UK over nine years (from 2009 to 2017), the bicyclist injury severity (with three severity levels: fatal injury, serious injury, and slight injury) was estimated using the generalized ordered logit (GOL) model and partial proportional odds (PPO) model. The marginal effects of each explanatory variable were computed to investigate the impacts on bicyclist injury severity occurring probabilities. A wide range of variables potentially affecting injury severity was considered, including bicyclist characteristics, intersection characteristics, environmental conditions, bicyclist movement and location preceding the crash, and types of collisions. Our findings show that the PPO model outperforms the GOL model for analyzing the factors that affect the bicyclist injury severity at intersections. The factors that affect cycling safety at various intersections show enormous differences. Specifically, nine variables have significant impacts on bicyclist injury severity at those three types of intersections. And there are only two variables, four variables, and eleven variables that have significant impact on bicyclist injury severity at roundabouts, crossroads, and T-junctions, respectively. The findings of this study can help decision makers better understand the spatial heterogeneity of the factors that influence the bicyclist injury severity at various intersections.

show abstract

“…Random heterogeneity treatments allow estimation of a vector of (safety event-specific)  parameter estimates on specific exogenous factors (e.g., volatility measures) by assuming a certain distribution in the population. Whereas, scale heterogeneity methods (as implemented in this study) capture the heterogeneous associations by a pure scale effect (i.e., across safety events, all  estimates are scaled up or down in tandem) -implying that mechanisms leading 1 In the context of the impacts of exogenous variables on driver injury severity using traditional General Estimates System (GES) database, a recent study carefully investigated whether the potential heterogeneous associations between exogeneous factors and injury severity could be better represented through a scaled or random heterogeneity treatment in an ordered discrete framework (Marcoux et al 2018). In doing so, a scaled ordered logit model was compared with a mixed (random parameter) ordered logit model concluding the statistical superiority of the earlier in terms of data fit -i.e., much of the heterogeneity in the associations can be captured by a pure scale effect.…”

Section: Research Gap Objectives and Contributionmentioning

confidence: 99%

“…In doing so, a scaled ordered logit model was compared with a mixed (random parameter) ordered logit model concluding the statistical superiority of the earlier in terms of data fit -i.e., much of the heterogeneity in the associations can be captured by a pure scale effect. However, as the study acknowledged, mixed generalized ordered logit and scaled generalized ordered logit models were estimated separately precluding a simultaneous examination of scale and random heterogeneity (Marcoux et al 2018).…”

Section: Research Gap Objectives and Contributionmentioning

confidence: 99%

Harnessing ambient sensing & naturalistic driving systems to understand links between driving volatility and crash propensity in school zones – A generalized hierarchical mixed logit framework

Wali

Khattak

2020

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

With the advent of seemingly unstructured big data, and through seamless integration of computation and physical components, cyber-physical systems (CPS) provide an innovative way to enhance safety and resiliency of transport infrastructure. This study focuses on real-world microscopic driving behavior and its relevance to school zone safety-expanding the capability, usability, and safety of dynamic physical systems through data analytics. Driving behavior and school zone safety is a public health concern. The sequence of instantaneous driving decisions and its variations prior to involvement in safety critical events, defined as driving volatility, can be a leading indicator of safety. By harnessing unique naturalistic data on more than 41,000 normal, crash, and near-crash events featuring over 9.4 million temporal samples of real-world driving, a characterization of volatility in microscopic driving decisions is sought at school and non-school zone locations. A big data analytic methodology is proposed for quantifying driving volatility in microscopic real-world driving decisions. Eight different volatility measures are then linked with detailed event-specific characteristics, health history, driving history/experience, and other factors to examine crash propensity at school zones. A comprehensive yet fully flexible state-of-the-art generalized mixed logit framework is employed to fully account for distinct yet related methodological issues of scale and random heterogeneity, containing multinomial logit, random parameter logit, scaled logit, hierarchical scaled logit, and hierarchical generalized mixed logit as special cases. The results reveal that both for school and non-school locations, drivers exhibited greater intentional volatility prior to safety-critical events. Evidence is found that an increase in volatility in positive and negative vehicular jerk in longitudinal and lateral direction increases the probability of unsafe outcomes (crashes or near-crashes) at school zones. A one-unit increase in intentional volatility associated with positive vehicular jerk in longitudinal direction increases the probability of crash outcome by 0.0528 units. Importantly, the effect of negative vehicular jerk (braking) in longitudinal direction on the likelihood of crash outcome is almost double. Methodologically, Hierarchical Generalized Mixed Logit model resulted in best-fit, simultaneously accounting for scale and random heterogeneity. When accounted for separately, more parsimonious models accounting for scale heterogeneity performed comparably to the less parsimonious counterparts accounting for random heterogeneity. Importantly, even after accounting for random heterogeneity, substantial heterogeneity due to a "pure scale-effect" is still observed, underscoring the importance of scale effects in influencing the overall contours of variations in modeled relationships. The study demonstrates the value of observational study design and big data analytics for understanding extreme driving behaviors in safe vs. unsafe ...

show abstract

Evaluating temporal variability of exogenous variable impacts over 25 years: An application of scaled generalized ordered logit model for driver injury severity

Cited by 40 publications

References 44 publications

Do high visibility crosswalks improve pedestrian safety? A correlated grouped random parameters approach using naturalistic driving study data

Do high visibility crosswalks improve pedestrian safety? A correlated grouped random parameters approach using naturalistic driving study data

Determinants of Bicyclist Injury Severity Resulting from Crashes at Roundabouts, Crossroads, and T-Junctions

Harnessing ambient sensing & naturalistic driving systems to understand links between driving volatility and crash propensity in school zones – A generalized hierarchical mixed logit framework

Contact Info

Product

Resources

About