EFFECTS OF PARAMETER SELECTION ON FORECAST ACCURACY AND EXECUTION TIME IN NONPARAMETRIC REGRESSION

In this paper, the authors discuss how nonparametric regression can be used to accurately forecast short term traffic flow. They first describe how nonparametric regression is a forecasting technique based on nearest neighbor searching in which forecasts are derived from previous observations that are similar to current conditions. They then discuss how the execution time of the nearest neighbor nonparametric regression can be achieved through advance data structures. They also offer that further reductions may be achieved by using approximate nearest neighbors, although this leads to a reduction forecast accuracy.

• Supplemental Notes:
  • Publication Date: 2000. IEEE Service Center, Piscataway NJ
• Corporate Authors:

  University of Dundee. Dept. of Applied Computing

  ,    

  Ohio State University, Columbus

  Department of Electrical Engineering
  Columbus, OH  United States  43210

  Technische Universiteit Eindhoven

  ,    

  Universidade de Sao Paulo

  ,    

  Honda Gijutsu Kenkyujo

  ,    

  Tsukuba Daigaku

  ,    
• Authors:
  • Smith, Brian Lee
  • Oswald, R Keith
• Conference:
  • IEEE Intelligent Transportation Systems Conference (2000 : Dearborn Mich.). 2000 IEEE intelligent transportation systems proceedings
• Publication Date: 2000

Language

• English

Media Info

• Pagination: p. 252-257

Subject/Index Terms

• TRT Terms: Traffic estimation

Filing Info

• Accession Number: 00961190
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Aug 4 2003 12:00AM