Ranking of Junctions in Directed Acyclic Graphs Inferred from Vehicle Traffic Time Series
Queen, Wright and Albers (2007) present a method to "elicit" an acyclic-directed graph (DAG) from a multivariate time series of vehicle counts in a vehicle road network. In other words, from a road network and data about vehicle traffic, one can get DAGs which can be exploited for a variety of purposes. In this research, the student will investigate the linear multiregression dynamic model (LMDM) by Queen, Wright and Albers (2007) to produce DAGs from vehicle data. Then, the student will apply node centrality algorithms, such as the gravity centrality index, to rank the nodes of the DAG with the objective of finding correlations between traffic bottlenecks and the centrality of nodes.
References
Queen, C. M., Wright, B. J., & Albers, C. J. (2007). Eliciting a directed acyclic graph for a multivariate time series of vehicle counts in a traffic network. Australian & New Zealand Journal of Statistics, 49(3), 221-239. https://doi.org/10.1111/j.1467-842X.2007.00477.x
Ma, L. L., Ma, C., Zhang, H. F., & Wang, B. H. (2016). Identifying influential spreaders in complex networks based on gravity formula. Physica A: Statistical Mechanics and its Applications, 451, 205-212.