METHODS OF LONG TERM FORECASTING AS EXEMPLIFIED BY THE DEGREE OF PASSENGER CAR MOTORISATION

METHODEN LANGFRISTIGER PROGNOSEN, GEZEIGT AM BEISPIEL DER PKW-MOTORISIERUNG

THIS PAPER DEALS WITH THE LONG TERM FORECASTING OF PASSENGER CAR MOTORISATION. THE SOLUTION USES THE METHOD OF NON-LINEAR REGRESSION (ON THE PRINCIPLE OF LEAST SQUARES), WHICH IS APPLIED TO 4 MODELS, TWO VARIANTS EACH OF THE GOMPERTZ FUNCTION AND THE LOGISTIC FUNCTION. DEGREES OF SATURATION OF THE DEVELOPMENT PROCESSES ARE NOT ASSUMED TO BE A PRIORI CONSTANTS, BUT ARE CALCULATED FROM THE REGRESSION. SPECIAL ATTENTION WAS DEVOTED TO GIVING A PROBABILITY DEPENDENT CONFIDENCE INTERVAL TO EACH FORECAST VALUE. THE METHOD HAS GENERAL VALIDITY. THE PRACTICAL APPLICABILITY IS DEMONSTRATED BY CARRYING OUT A FORECAST FOR PASSENGER CARS IN THE FEDERAL REPUBLIC OF GERMANY (BOTH FOR THE COUNTRY AS A WHOLE AND SEPARATELY FOR EACH FEDERAL STATE). THE COMPUTATION WAS CARRIED OUT WITH THE AID OF AN ELECTRONIC COMPUTER AND IS PROGRAMMED IN SUCH A FASHION THAT ANY MACHINE HAVING A FORTRAN IV COMPILER CAN BE USED, AND PERMITS VARIATION OF THE ACCURACY OF THE ITERATION AND CONFIDENCE LIMITS AND ALSO EXTRAPOLATION WITHOUT ALTERATION TO THE PROGRAMS. THE PROGRAMS MAY ALSO BE USED FOR FORECASTS RELATING TO OTHER AREAS.

Language

  • German

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01303270
  • Record Type: Publication
  • Source Agency: Forschungsgesellschaft für Straßen- und Verkehrswesen (FGSV)
  • Files: ITRD
  • Created Date: Nov 21 2010 9:00AM