A Correlated Random Parameters Model with Heterogeneity in Means to Account for Unobserved Heterogeneity in Crash Frequency Analysis

Abstract

Unobserved heterogeneity induced by omitted variables is a major challenge in developing reliable road safety models. In recent years, the random parameters negative binomial (RPNB) model has been used frequently in crash frequency analysis to account for unobserved heterogeneity. However, the majority of past studies of the RPNB model assumed that there was no correlation between different sources of unobserved heterogeneity, which is not always true given the complex interactions of safety factors. Compared with the RPNB model, a more flexible random parameters model that is the correlated random parameters negative binomial with heterogeneity in means (CRPNBHM) model was proposed in this study. Results indicate that the CRPNBHM model could not only capture the otherwise unobserved heterogeneity, but also track the underlying correlation among different sources of unobserved heterogeneity, thus outperforming the RPNB model. In addition, new insights into the interactions of safety factors (e.g., the joint safety effects of heavy trucks and pavement rutting depth) were obtained from the CRPNBHM model and these are expected to be beneficial in developing effective safety countermeasures. Results from this study demonstrated the CRPNBHM model to be a good alternative for crash frequency analysis, particularly when unobserved heterogeneity was detected.

Application of the Apriori Algorithm for Traffic Crash Analysis in Thailand

“…Thailand has over 42 million registered vehicles, with motorcycles accounting for 50% of the total and being responsible for the majority of road fatalities [58], thus potentially contributing to the largest number of fatalities from significant accidents. As Jomnonkwao, Uttra and Ratanavaraha [2] observed, motorcyclists are responsible for the vast majority of road fatalities; meanwhile, prior studies showed different types of cars and motorcycles, such as SUV/vans [47], large trucks [40,47,49], and pick-ups [40], that have rolled over. Additional research on motorcycle riders specifically, as well as other types of road users, may be conducted in the future.…”

Analysis of the Effects of Highway Geometric Design Features on the Frequency of Truck-Involved Rear-End Crashes Using the Random Effect Zero-Inflated Negative Binomial Regression Model

“…Islam, et al [35] highlighted that curved roads necessitate a reduction in speed, particularly in cases of high traffic volume, increasing the risk of rear-end collisions. Slope variation is another factor influencing the frequency of rear-end collisions, especially for trucks, due to brake-related issues when repeatedly applying the brakes [36]. Finally, the presence of median openings also contributes to the likelihood of rear-end collisions, as observed by Champahom, et al [37].…”

Determinants and Prediction of Injury Severities in Multi-Vehicle-Involved Crashes

“…Many previous studies emphasized the importance of taking the unobserved heterogeneity into consideration while analyzing crash-related factors. Furthermore, random parameters logit (RPL) models [ 16 , 17 , 18 , 19 ], random parameters logit models with heterogeneities in the means [ 20 , 21 ], and random parameters logit models with heterogeneities in the means and variances [ 5 , 22 , 23 ] have all been successfully utilized in the investigation of crash injury-severity. Hence, to account for the unobserved heterogeneity, the random parameters logit model with heterogeneities in the means and variances was used in this study.…”