ON SOME INFORMATION THEORETICAL ASPECTS OF THE ENTROPY MAXIMISING METHOD

OVER ENKELE INFORMATIETHEORETISCHE ASPEKTEN VAN DE METHODE VAN ENTROPIE-MAXIMALISATIE

Comments are made on the entropy optimising method, in particular on the objective function of this principle, named the entropy function. Firstly aspects of the statistical information theory of this method are discussed. Secondly, attention is paid to the conditions that govern the valid use of the objective function. Two different ways of defining the objective function are discussed. The first is based on the number of possible combination of individuals that leads to the same spatial distribution. The second is based on the probability function of Bernoulli. The conditions to which both objective functions are liable are very restrictive. It is concluded that perhaps it would be preferable to replace the entropy optimising method by the maximum likelihood method. /TRRL/

  • Corporate Authors:

    Colloguium Vervoersplanologisch Speurwerk

    P.O. Box 45
    Delft,   Netherlands 
  • Authors:
    • Pluym, W K
  • Publication Date: 1974

Language

  • Dutch

Media Info

  • Features: References; Tables;
  • Pagination: 12 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00126096
  • Record Type: Publication
  • Source Agency: Transport and Road Research Laboratory (TRRL)
  • Report/Paper Numbers: Conf Paper
  • Files: ITRD, TRIS
  • Created Date: May 14 1976 12:00AM