THE APPLICATION OF AN ELIMINATION-BY-ASPECTS MODEL TO RESIDENTIAL LOCATION CHOICE

The choice of residential location is characterised by the existence of many alternatives, each described in terms of many attributes. In such a complex choice situation, it is entirely plausible that a decision-maker may attempt to simplify the choice such that a decision is made before all attributes have been considered. This paper describes the development of a model which explicitly accounts for this type of behaviour. The theory behind the model, known as an elimination-by-aspects model, is discussed and its extension to allow for use of maximum likelihood estimation procedures is described. The model is then calibrated using a data set of 716 observations of residential location behaviour in Melbourne. Statistical testing of the model and its estimated tolerances shows that the model explains a high proportion of variance in the data set. Split sample model calibration and cross-prediction demonstrates high stability in the calibrated models and tolerances. Finally, choice elasticities are calculated and are shown to yield highly non-linear effects with different percentage changes in attribute satisfaction. It is concluded that the model should, with continued development, provide a useful input to informed policy advice in the area of residential location choice. (Author/TRRL)

  • Availability:
  • Corporate Authors:

    Monash University

    Wellington Road
    Clayton, Victoria  Australia  3800
  • Authors:
    • YOUNG, W
    • Richardson, A J
  • Publication Date: 1981-7

Media Info

  • Features: Figures; References; Tables;
  • Pagination: 25 p.
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00361630
  • Record Type: Publication
  • Source Agency: ARRB
  • Report/Paper Numbers: Monograph
  • Files: ITRD, TRIS, ATRI
  • Created Date: Jul 30 1982 12:00AM