Comparative Evaluation of The Traffic Flow Volatility Forecasting Models
Jiawei Lu, Hongjun Xue, Guangjiao Chen, You Zhou, Jingxin Xia
Available Online October 2016.
- https://doi.org/10.2991/ceie-16.2017.44How to use a DOI?
- Traffic Flow Volatility; Short-term Forecasting; Performance Evaluation; Comparative Analysis
- Accurate and reliable short-term traffic flow forecasting is essential for advanced traveler information systems and proactive traffic signal control systems. However, the majority of current studies mainly concentrate on short-term traffic flow level forecasting. To improve the forecasting reliability of traffic flow parameters, short-term traffic flow uncertainty forecasting has been emphasized and studied gradually, and many models had been developed. Among these models, generalized autoregressive conditional heteroscedasticity (GARCH) and stochastic volatility (SV) model is considered to have high accuracy and reliability. Based on traffic flow data collected from urban roads, forecasting methods of forecasting performance of short-term traffic flow uncertainty are studied and compared under different roads and traffic conditions in this paper. The results show that the performance of the ARIMA-GARCH, VAR-MGARCH, and ARIMA-SV models for short-term speed forecasting is better than the performance for volume forecasting. Moreover, the accuracy performance of VAR-MGARCH model is better than the ARIMA-SV and ARIMA-GARCH models, but the ARIMA-SV model performs better than the ARIMA-GARCH and VAR-MGARCH models in terms of the short-term uncertainty forecasting.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Jiawei Lu AU - Hongjun Xue AU - Guangjiao Chen AU - You Zhou AU - Jingxin Xia PY - 2016/10 DA - 2016/10 TI - Comparative Evaluation of The Traffic Flow Volatility Forecasting Models BT - Proceedings of the International Conference on Communication and Electronic Information Engineering (CEIE 2016) PB - Atlantis Press SP - 348 EP - 363 SN - 2352-5401 UR - https://doi.org/10.2991/ceie-16.2017.44 DO - https://doi.org/10.2991/ceie-16.2017.44 ID - Lu2016/10 ER -