THE EFFECTS OF MEASUREMENT TECHNIQUES, VARIABLE DEFINITION AND MODEL SPECIFICATION ON DISAGGREGATE DEMAND MODEL FUNCTIONS

A fine disaggregate data base consisting of some 700 individuals travelling to work in an urban corridor of Santiago was collected in 1981. The data consider level-of-service variables (reported and measured) and socio-economic variables (including income) for every individual. Part of the data (217 individuals) was coded in two different ways. Firstly, considering only those alternative modes mentioned by the individuals and secondly, adding to the choice set those modes which were obvious (although not reported) alternatives. The paper considers in its first part the effects on model coefficients and structure of estimating the demand functions with both data sets. In its second part the paper examines the effects of estimating the demand functions with the complete sample and with a sub sample of some 220 individuals who have exactly the same choice set (they reside in a subcorridor close to the underground line). (Author/TRRL)

  • Availability:
  • Supplemental Notes:
    • This paper was presented at Seminar N, Developing Countires, held at the PTRC Annual Summer Meeting, University of Sussex, England, 4-7 July, 1983.
  • Corporate Authors:

    PTRC Education and Research Services Limited

    110 Strand
    London WC2,   England 
  • Authors:
    • Deown, H P
    • DONOSO, PCF
    • HUTT, G A
  • Conference:
  • Publication Date: 1983

Media Info

  • Features: Figures; References; Tables;
  • Pagination: p. 299-310

Subject/Index Terms

Filing Info

  • Accession Number: 00382866
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • ISBN: 0-86050-112-4
  • Report/Paper Numbers: Volume P240
  • Files: ITRD, TRIS
  • Created Date: May 30 1984 12:00AM