SHORT-TERM FORECASTING OF GASOLINE DEMAND

Techniques used recently by the U.S. Department of Energy to forecast short term demand for motor-vehicle gasoline are reviewed. Techniques used during and before 1979 are discussed briefly, and the rationale for the development of new methods during 1980 is also presented. Because the forecasting effort is an ongoing one, the procedures evolve over time. Only the techniques developed during 1980 are treated in detail, but a brief discussion and summary of the older methods are provided for comparison purposes. The current forecasting technique relies on predetermined parameter values rather than econometrically estimated values. This is the result of an evaluation of the econometric estimates. The new procedures have resulted in improved forecast accuracy and have anticipated the downturn in motor-vehicle gasoline demand that occurred in 1980. The current model computes annual demand for 1980 within 1.0 percent of actual demand, and the average error for the monthly demand estimates during 1980 is less than 2.5 percent of actual demand. The current techniques can be used to project the effects of various policy options, such as improved mileage requirements or gasoline tax levies. (Authors)

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: pp 22-28
  • Monograph Title: Energy: forecasting, data and conservation
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00342120
  • Record Type: Publication
  • ISBN: 0309032156
  • Files: TRIS, TRB
  • Created Date: Jan 29 1982 12:00AM