Traffic Volatility Forecasting for Urban Arterials: Family GARCH Approach

Traffic flow volatility modeling has been highly valued in recent years because of its advantages to describe the uncertainty of traffic flow during the short-term forecasting process. Some heteroscedasticity models and the stochastic volatility model have been proposed to capture and hence to forecast the volatility of traffic variables, which were confirmed to be able to produce more accurate and reliable forecasts than the conditional level forecasting models. However, few of these researches can fully capture the vital attributes of traffic flow volatility, such as volatility clustering, Taylor effect, long memory and leverage effect. Therefore, an auto regression integrated moving average (ARIMA) method plus family generalized autoregressive conditional heteroscedasticity (FGARCH) method is proposed for short-term traffic flow volatility forecasting model the volatility attributes that have been ignored in the past. The ARIMA model is selected as the mean equation of FGARCH model for traffic flow level forecasting, while the FGARCH model is used to forecast the traffic flow volatility. Actual traffic speed data collected from the Kunshan city in China was used for model validation and evaluation. The empirical results show that the proposed method outperforms the method of ARIMA plus GARCH in the aspect of both level and reliability forecasting, especially under congested traffic state.

• Supplemental Notes:
  • This paper was sponsored by TRB committee AHB45 Traffic Flow Theory and Characteristics.
• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Zhou, Yang
  • Nie, Qinghui
  • Xia, Jingxin
  • An, Chengchuan
• Conference:
  • Transportation Research Board 94th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2015-1-11 to 2015-1-15
• Date: 2015

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 13p
• Monograph Title: TRB 94th Annual Meeting Compendium of Papers

Subject/Index Terms

• TRT Terms: Arterial highways; Heteroscedasticity; Traffic flow; Traffic forecasting; Traffic speed; Urban highways
• Uncontrolled Terms: Generalized autoregressive conditional heteroscedasticity
• Geographic Terms: Kunshan City (China)
• Subject Areas: Highways; Planning and Forecasting; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01557321
• Record Type: Publication
• Report/Paper Numbers: 15-3618
• Files: TRIS, TRB, ATRI
• Created Date: Mar 19 2015 12:06PM