Application of Subset Autoregressive Integrated Moving Average Model for Short-Term Freeway Traffic Volume Forecasting

Abstract

Traffic volume is one of the fundamental types of data that have been used for the traffic control and planning process. Forecasting needs and efforts for various applications will be increased with the deployment of advanced traffic management systems. With the importance of the short-term traffic forecasting task, numerous techniques have been utilized to improve its accuracy. The use of the subset autoregressive integrated moving average (ARIMA) model for short-term traffic volume forecasting is investigated. A typical time-series modeling procedure was employed for this study. Model identification was carried out with Akaike’s information criterion. The conditional maximum likelihood method was used for the parameter estimation process. Two white noise tests were applied for model verification. From the analysis results, four time-series models in different categories were identified and used for the one-step-ahead forecasting task. The performance of each model was evaluated using two statistical error estimates. Results showed that all time-series models performed well with reasonable accuracy. However, it was observed that the subset ARIMA model gave more stable and accurate results than other time-series models, especially a full ARIMA model. It is believed that the use of a subset ARIMA model could increase the accuracy of the short-term forecasting task within time-series models.

Utilizing Real-World Transportation Data for Accurate Traffic Prediction

“…In the past, several statistics, machine learning and data mining approaches have been applied to traffic data for prediction purposes, such as auto-regression [12], neural net [21] and smoothing [22] techniques. However, in this paper, we took a very pragmatic approach to evaluate and then enhance these techniques by intensely studying a very largescale and high-resolution spatiotemporal transportation data from LA County road network.…”

Spatiotemporal Traffic State Prediction Based on Discriminatively Pre-trained Deep Neural Networks

“…Voort et al integrated the Kohonen self-organizing map and ARIMA into a new method called KARIMA, which uses a Kohonen self-organizing map to cluster the data and then models each cluster using ARIMA [6]. Lee et al used a subset ARIMA model for the one-step-ahead forecasting task, which returned more stable and accurate results than the full ARIMA model [7]. Williams and Hoel used seasonal ARIMA (SARIMA) to analyze data from two freeways, and the results showed that one-step seasonal ARIMA predictions outperformed heuristic forecast benchmarks [8].…”

How to Provide Accurate and Robust Traffic Forecasts Practically?

“…Thereinto, the autoregressive integrated moving average (ARIMA) (Ahmed & Cook, 1979) family of models such as simple ARIMA (Levin & Tsao, 1980;Nihan & Holmesland, 1980;Hamed et al, 1995;Smith, 1995;Williams, 1999), ATHENA , subset ARIMA (Lee & Fambro, 1999), SARIMA family (Smith et al, 2002;Williams et al, 1998Williams et al, , 2003Ghosh et al, 2005), are classical milestones in forecasting area. Such time series methods belong to time domain approaches, and frequency domain approaches like spectral analysis, "which are regressions on periodic sines and cosines, show their important insights into traffic data which may not apparent in an analysis in the time domain only" (Stathopoulos & Karlaftis, 2001a, b).…”