Testing for Aggregation Effects in Short-Term Traffic Flow Time-Series

Time series data are an invaluable tool in studying transportation phenomena. An issue that is often ignored in transportation time series analysis practice is the effect of temporal aggregation bias. Aggregation has direct implications to the temporal structure of time series, because it leads to “smother” series or alters many important properties that exist in the disaggregated frequency. However, many researchers suggest that data should be aggregated to avoid the effects of noise and spurious oscillations. We analyze the effect of data aggregation on traffic volume and occupancy series. Results indicate that aggregation may suppress the long memory characteristics of traffic flow.

Language

  • English

Media Info

  • Media Type: DVD
  • Features: Figures; References; Tables;
  • Pagination: 11p
  • Monograph Title: TRB 89th Annual Meeting Compendium of Papers DVD

Subject/Index Terms

Filing Info

  • Accession Number: 01152546
  • Record Type: Publication
  • Report/Paper Numbers: 10-0401
  • Files: BTRIS, TRIS, TRB
  • Created Date: Jan 25 2010 10:13AM