Generalized Estimating Equations Model Based Recursive Partitioning: Applied to Distracted Driving

Abstract

Conventional statistical methods have utilized a coarse aggregation of information crosswise over subjects that may not be illustrative of any single person. Despite the fact that Generalized Estimating Equations system stretches out summed up direct model to take into account examination of repeated estimations or other similar observations, the non-linear connection between dependent and independent variables that could hinder model performance. In this study, we propose Generalized Estimating Equation based tree model that combines the benefits of both models by separating the data set recursively into subsets with different parameter estimates. For the best use of the proposed model, distracted driving on the intersection is examined. Past examinations have concentrated on evaluating the singular effect of individual geometry and human characteristic variables on driving behaviors. Interactions between factors related with red-light running (e.g., cell phone use, cell phone interface, and driver age) introduce diverse levels of distraction on red-light running. We recognize interactions that are sensitive to red-light running and different as a function of the level of the speed and yellow interval duration. Drivers are more vulnerable to cell phone distractions when their location is close the stop line of the intersection at the beginning of yellow interval.