Unobserved Component Model for Predicting Monthly Traffic Volume

Traffic volume prediction plays a critical role in transportation system and infrastructure management. This paper develops the first application of an unobserved component model (UCM) for monthly traffic volume forecasting. We compare the UCM model with simple linear regression, autoregressive integrated moving average (ARIMA), support vector machine (SVM), and artificial neural network (ANN) models based on monthly traffic volume data from a key corridor in New Jersey. As a general econometric method, the UCM decomposes the time series into trend, seasonal, and irregular components, exhibiting superiority for statistically modeling traffic data with cyclic or seasonal fluctuations. The numerical analysis shows that the UCM outperforms all of the other four models and generates reasonably accurate prediction results. This research indicates that UCM can be considered as an alternative approach to modeling traffic volumes.