Grouped Random Parameters Negative Binomial-Lindley for accounting unobserved heterogeneity in crash data with preponderant zero observations

Islam, A.S.M. Mohaiminul; Shirazi, Mohammadali; Lord, Dominique

doi:10.1016/j.amar.2022.100255

Cited by 15 publications

(2 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study concluded that there were unobserved factors between segments or intersections in the same traffic analysis zone. A more recent study proposed a grouped random parameter negative binomial-Lindley model (G-RPNB-L) to account for unobserved heterogeneity in crash counts with a high percentage of zero occurrences [ 28 ]. The study applied the proposed model to lane departure crashes from rural interstate segments and compared it with standard negative binomial (NB), negative binomial-Lindley (NB-L), and grouped random parameter negative binomial (G-RPNB).…”

Section: Introductionmentioning

confidence: 99%

A nested grouped random parameter negative binomial model for modeling segment-level crash counts

Almutairi

2024

Heliyon

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A nested grouped random parameter negative binomial model for modeling segment-level crash counts

Almutairi

2024

Heliyon

View full text Add to dashboard Cite

“…Observed variables extracted from historical crash data were mainly utilized to estimate the direct effect of indicator variables on the severity of crashes involving a truck. Despite the selection of the model and its underlying assumptions, the complex interrelationship between crash variables cannot be observed using traditional parametric approaches ( 4 , 5 ). Unlike other studies, a comprehensive analysis framework was conducted in this study, in which parametric and latent factor analysis approaches were employed.…”

mentioning

confidence: 99%

Analysis Framework to Assess Crash Severity for Large Trucks on Rural Interstate Roads Utilizing the Latent Class and Random Parameter Model

Gaweesh

Ahmed²,

Ahmed

2023

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Truck freight has high importance to the national economy as it handles more cargo than other types of freight transportation. To improve the safety of commercial vehicles, several studies have focused on determining factors leading to crashes. Parametric models have been extensively employed to explain crash causal factors for heavy trucks. Unlike studies in the past, a comprehensive framework was proposed in this study to compare crash underlying factors utilizing several statistical approaches. The structural equation modeling approach was used to assess latent factors affecting the crash severity of large trucks. In addition, ordinary (binary) logistic and random parameter models were employed to assess the direct effect of the observed data, and the heterogeneity in parameter means was also estimated. Three crash categories were investigated: single-truck crashes, multi-vehicle truck crashes, and total truck crashes. A total of five years of crash data from 2015 to 2019 were analyzed. The results showed that multiple observed variables were factored to measure crash severity. Direct and indirect effects were identified, in which challenging roadway conditions had an indirect effect on crash severity for single-truck crashes. Random parameter logit models indicated that roadway geometry and adverse weather conditions were among the significant contributing factors increasing crash severity for trucks. This study recommends that improving the situational awareness of truck drivers, providing more frequent rest stops, updating variable speed limit algorithms, and integrating roadway geometry information into connected vehicle applications in Wyoming could be considered to assist stakeholders in promoting safety on rural interstate corridors.

show abstract

Assessing the Negative Binomial-Lindley model for crash hotspot identification: Insights from Monte Carlo simulation analysis

Gil-Marin,

Shirazi,

Ivan

2024

Accident Analysis & Prevention

View full text Add to dashboard Cite

Grouped Random Parameters Negative Binomial-Lindley for accounting unobserved heterogeneity in crash data with preponderant zero observations

Cited by 15 publications

References 57 publications

A nested grouped random parameter negative binomial model for modeling segment-level crash counts

A nested grouped random parameter negative binomial model for modeling segment-level crash counts

Analysis Framework to Assess Crash Severity for Large Trucks on Rural Interstate Roads Utilizing the Latent Class and Random Parameter Model

Assessing the Negative Binomial-Lindley model for crash hotspot identification: Insights from Monte Carlo simulation analysis

Contact Info

Product

Resources

About