A structural equations approach for modeling the endogeneity of lane-mean speeds considering the downstream speeds
Like other transportation data, lane-mean speeds are also best modeled by a system of structural equations. Several studies omit the interrelation between adjacent lane speeds, which may produce biased and inconsistent results if models are solved by ordinary least squares. The uncorrelatedness of regressors and disturbances assumption of ordinary least squares is violated since one or more independent variables are endogenous in the system. This study attempts to propose a structural equations approach to model the lane-mean speeds in multi-lane traffic, in which the endogeneity of adjacent lane speeds and the downstream speeds are being considered. Additionally, the equations system can serve as a prediction model for lane-mean speeds. Several empirical analyses using the data collected from multi-lane freeways with different lengths and different numbers of lanes are conducted to observe the performance of the equations system in different conditions. The study further compares the prediction accuracy between the underlying approach and the model established by Shankar and Mannering (1998) for assessing the impact of introducing downstream speeds within the model. The findings show that more precise results are obtained generally after downstream speeds are included, emphasizing the improvements and superiority of this approach.
This work is licensed under a Creative Commons Attribution 4.0 International License.
EJTIR is licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence. The license means that anyone is free to share (to copy, distribute, and transmit the work), to remix (to adapt the work) under the following conditions:
- The original authors must be given credit
- For any reuse or distribution, it must be made clear to others what the license terms of this work are
- Any of these conditions can be waived if the copyright holders give permission
- Nothing in this license impairs or restricts the author's moral rights