1,721,060 research outputs found
Computational Bayesian statistics in transportation modeling: from road safety analysis to discrete choice
No full textIn this paper, we review both the fundamentals and the expansion of computational Bayesian econometrics and statistics applied to transportation modeling problems in road safety analysis and travel behavior. Whereas for analyzing accident risk in transportation networks there has been a significant increase in the application of hierarchical Bayes methods, in transportation choice modeling, the use of Bayes estimators is rather scarce. We thus provide a general discussion of the benefits of using Bayesian Markov chain Monte Carlo methods to simulate answers to the problems of point and interval estimation and forecasting, including the use of the simulated posterior for building predictive distributions and constructing credible intervals for measures such as the value of time. Although there is the general idea that going Bayesian is just another way of finding an equivalent to frequentist results, in practice Bayes estimators have the potential of outperforming frequentist estimators and, at the same time, may offer more information. Additionally, Bayesian inference is particularly interesting for small samples and weakly identified models.</p
A mixed-effect statistical model for before-after speed studies
No full textThis paper proposes an efficient methodology to conduct observational before-after studies for operating speed data. The employed method has some noteworthy strengths: (i) it can analyze data at a disaggregated level to properly account for variations in the speed profile; (ii) it considers the entire distribution of speed to overcome the bias associated to the traditional approaches that represent the speed distribution with a single point estimate; (iii) it takes advantage of full Bayes methods to avoid the empirical Bayes method limitations. To illustrate the feasibility of the proposed framework, a limited sample of before-after speed dataset from Montreal was used. The effectiveness of a safety countermeasure-a reduction in speed limits-was assessed. The speed data were collected on local urban streets grouped into comparison and treatment sites. For modeling the operating speed, we employed a hierarchical mixed-effect Binomial model using a wide range of site characteristics. This model is capable of dealing with heterogeneity across observations and accounting for site specific effects. The analyses results indicated that lane width, number of lanes, and night hours affect the operating speed positively while presence of parking, peak hours, weekend, one way, and precipitation affect it negatively. Although the speed limit reduction was found to be effective in controlling the operating speed, the analyzed sample may not be representative of the entire urban areas subject to this reduction. This paper also highlights some essential issues in the data collection process and the sensitivity of the outcomes to the collection method used.</p
Does prior specification matter in hotspot identification and before-after road safety studies?
No full textThis study investigated the effects of prior assumptions in applications of full Bayes methods in road safety analysis. The effect of prior choice was evaluated in the accuracy of model parameters, hotspot identification, goodness of fit, and a treatment effectiveness index in before-after studies. Particular attention was devoted to conditions with a lack of data, which were referenced as the low-mean and small-sample problem. In this research, informative, semi-informative, and noninformative priors were determined on the basis of past published studies. A simulation framework was used to evaluate various scenarios of sample size and crash occurrence mean. Quasi-simulated data were generated on the basis of two empirical databases of divided and undivided rural highway segments in New York and Texas. Various sample mean values were obtained on the basis of time period (number of years) and classifying accidents in fatal-injury and total accidents. The outcomes under low-mean and small-sample conditions were found to be significantly biased. However, the introduction of informative priors can make observational before-after studies feasible when a few observations from treatment or comparison sites are used. Informative priors can help provide more accurate estimates of the effectiveness of the treatment. Finally, in accordance with previous works, the inverse dispersion parameter was significantly affected by prior specifications; nevertheless, regression parameters, goodness of fit, and hotspot identification were less sensitive to prior choices.</p
On the causal effect of proximity to school on pedestrian safety at signalized intersections: a heterogeneous endogenous econometric model
No full textPedestrian safety in proximity to schools is a major concern of transportation authorities, local governments, and residents. In fact, several countermeasures (e.g., school-zone speed limits) are usually in place around schools to provide a safer environment, especially for school-age children. Two questions arise here: (i) are transportation facilities in proximity to schools truly safer than other facilities given a variety of implemented road safety interventions around schools? and (ii) how can we answer the previous question properly using a reliable approach that accounts for possible confounding? While previous literature has mixed results and does not provide clear methodological/empirical guidelines in this regard, we propose an approach that answers the above questions. We illustrate our method on a sample of intersections in Montreal, Canada. Specifically, to underpin a causal interpretation, for the first time in the extent of transportation literature, we develop a heterogeneous endogenous econometric model that estimates the causal effect of proximity to school on pedestrian safety, addressing a complex endogenous relationship between the two. Various built environment, traffic exposure, and road geometric/operational characteristics are considered. The results indicate that if endogeneity is not accounted for, the effect of proximity to school is underestimated, while not being significant at a 5% level of significance. However, after accounting for confounding factors, the proposed endogenous model indicates that proximity to school deteriorates pedestrian safety. Therefore, traffic safety countermeasures and policies in place (if any) during the study period have not been sufficient and/or effective in improving pedestrian safety at intersections near schools. Our heterogeneity in mean and variance formulation provided more insights. For example, we found that, interestingly, as pedestrian volume increases at intersections around schools, the adverse effect of proximity to school on pedestrian safety decreases, a possibility not previously explored in the extent of road safety literature, confirming a strong safety-in-numbers effect
Bayesian nonparametric modeling in transportation safety studies: Applications in univariate and multivariate settings
No full textIn transportation safety studies, it is often necessary to account for unobserved heterogeneity and multimodality in data. The commonly used standard or over-dispersed generalized linear models (e.g., negative binomial models) do not fully address unobserved heterogeneity, assuming that crash frequencies follow unimodal exponential families of distributions. This paper employs Bayesian nonparametric Dirichlet process mixture models demonstrating some of their major advantages in transportation safety studies. We examine the performance of the proposed approach using both simulated and real data. We compare the proposed model with other models commonly used in road safety literature including the Poisson-Gamma, random effects, and conventional latent class models. We use pseudo Bayes factors as the goodness-of-fit measure, and also examine the performance of the proposed model in terms of replicating datasets with high proportions of zero crashes. In a multivariate setting, we extend the standard multivariate Poisson-lognormal model to a more flexible Dirichlet process mixture multivariate model. We allow for interdependence between outcomes through a nonparametric random effects density. Finally, we demonstrate how the robustness to parametric distributional assumptions (usually the multivariate normal density) can be examined using a mixture of points model when different (multivariate) outcomes are modeled jointly.</p
Bayesian methodology to estimate and update safety performance functions under limited data conditions: A sensitivity analysis
No full textIn road safety studies, decision makers must often cope with limited data conditions. In such circumstances, the maximum likelihood estimation (MLE), which relies on asymptotic theory, is unreliable and prone to bias. Moreover, it has been reported in the literature that (a) Bayesian estimates might be significantly biased when using non-informative prior distributions under limited data conditions, and that (b) the calibration of limited data is plausible when existing evidence in the form of proper priors is introduced into analyses. Although the Highway Safety Manual (2010) (HSM) and other research studies provide calibration and updating procedures, the data requirements can be very taxing. This paper presents a practical and sound Bayesian method to estimate and/or update safety performance function (SPF) parameters combining the information available from limited data with the SPF parameters reported in the HSM. The proposed Bayesian updating approach has the advantage of requiring fewer observations to get reliable estimates. This paper documents this procedure. The adopted technique is validated by conducting a sensitivity analysis through an extensive simulation study with 15 different models, which include various prior combinations. This sensitivity analysis contributes to our understanding of the comparative aspects of a large number of prior distributions. Furthermore, the proposed method contributes to unification of the Bayesian updating process for SPFs. The results demonstrate the accuracy of the developed methodology. Therefore, the suggested approach offers considerable promise as a methodological tool to estimate and/or update baseline SPFs and to evaluate the efficacy of road safety countermeasures under limited data conditions.</p
Using a flexible multivariate latent class approach to model correlated outcomes: A joint analysis of pedestrian and cyclist injuries
No full textSeveral recent transportation safety studies have indicated the importance of accounting for correlated outcomes, for example, among different crash types, including differing injury-severity levels. In this paper, we discuss inference for such data by introducing a flexible Bayesian multivariate model. In particular, we use a Dirichlet process mixture to keep the dependence structure unconstrained, relaxing the usual homogeneity assumptions. The resulting model collapses into a latent class multivariate model that is in the form of a flexible mixture of multivariate normal densities for which the number of mixtures (latent components) not only can be large but also can be inferred from the data as part of the analysis. Therefore, besides accounting for correlation among crash types through a heterogeneous correlation structure, the proposed model helps address unobserved heterogeneity through its latent class component. To our knowledge, this is the first study to propose and apply such a model in the transportation literature. We use the model to investigate the effects of various factors such as built environment characteristics on pedestrian and cyclist injury counts at signalized intersections in Montreal, modeling both outcomes simultaneously. We demonstrate that the homogeneity assumption of the standard multivariate model does not hold for the dataset used in this study. Consequently, we show how such a spurious assumption affects predictive performance of the model and the interpretation of the variables based on marginal effects. Our flexible model better captures the underlying complex structure of the correlated data, resulting in a more accurate model that contributes to a better understanding of safety correlates of non-motorist road users. This in turn helps decision-makers in selecting more appropriate countermeasures targeting vulnerable road users, promoting the mobility and safety of active modes of transportation.</p
- …
