Spatial Correlation in Multilevel Crash Frequency Models

Abstract

Recent research has shown the importance of spatial correlation in road crash models. Because many different spatial correlation structures are possible, however, this study tested several segment neighboring structures to establish the most promising one to model crash frequency in road networks. A multilevel approach was also used to account for the spatial correlation between road segments of different functional types, which are usually analyzed separately. The study employed a full Bayes hierarchical approach with conditional autoregressive effects for the spatial correlation terms. Analyses of crash, traffic, and roadway inventory data from rural engineering districts in Pennsylvania and Washington affirmed the importance of spatial correlation in road crash models. Pure distance-based neighboring models (i.e., exponential decay) performed poorly compared with adjacency-based or distance order models. The results also suggest that spatial correlation is more important in distances of 1 mi or less. The inclusion of spatially correlated random effects significantly improved the precision of the estimates of the expected crash frequency for all segments by pooling strength from their neighbors and thus reducing their standard deviation. Results from Pennsylvania and Washington showed that spatial correlation substantially increased the random effects. There was a consistent indication that 70% to 90% of the variation explained by the random effects resulted from spatial correlation. This suggests that spatial models offer a significant advantage, since poor estimates that result from small sample sizes and low sample means are a frequent issue in highway safety analysis. Application of spatial correlation to the identification of sites with promise indicated that more sites were identified because of a reduction in the variance of the estimates, which would allow for greater confidence in the selection of sites for treatment.

Support vector machine in crash prediction at the level of traffic analysis zones: Assessing the spatial proximity effects

“…Specifically, to gain a more precise estimation of the variability in parameters by considering more complex spatial proximity structures, researchers have proposed a comprehensive investigation of different spatially neighboring structures for both roadsegment-level and area-wide analyses (i.e., Aguero-Valverde and Jovanis, 2010;Wang et al, 2012). As in the study by Dong et al (2014), CPMs accounting for spatial correlation perform better than non-spatial model and also model merely considering 0-1 first order adjacency-based proximity structure.…”

A Full Bayesian multivariate count data model of collision severity with spatial correlation

“…Congdon (2006) suggested that ignoring spatial dependence leads to an underestimation of variability. Furthermore, according to Aguero-Valverde and Jovanis (2010), the inclusion of spatially correlated random effects significantly improves the precision of the estimates of the expected collision frequency for road segments. The inclusion of spatial correlation has two main advantages: i) spatial correlation sites estimate the "pool strength" from neighboring sites, thereby improving model parameter estimation (Aguero-Valverde and Jovanis, 2008); and ii) spatial dependence can be a surrogate for unknown and relevant covariates, thereby reflecting unmeasured confounding factors (Cressie, 1993;Dubin, 1988).…”

The Effects of Traffic Composition on Freeway Crash Frequency by Injury Severity: A Bayesian Multivariate Spatial Modeling Approach

“…Proximity structure is a critical element of the CAR prior. While various proximity structures have been investigated [32], the most prevalent structure, 0-1 first-order neighbor, is used to define the proximity matrix in this study. Specifically, if segments and are connected to each other directly, , = 1; otherwise, , = 0.…”