Premium
The Statistical Modeling of Flow Data When the Poisson Assumption Is Violated
Author(s) -
Davies Richard B.,
Guy Clifford M.
Publication year - 1987
Publication title -
geographical analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.773
H-Index - 65
eISSN - 1538-4632
pISSN - 0016-7363
DOI - 10.1111/j.1538-4632.1987.tb00132.x
Subject(s) - operationalization , poisson distribution , computer science , statistical model , flow (mathematics) , econometrics , mathematics , statistics , artificial intelligence , physics , geometry , quantum mechanics
The Poisson model typically provides a poor fit to flow data but more complex models are difficult to operationalize, especially with production or attraction constraints. Quasi‐ and pseudo‐likelihood approaches retain the attractive computational features of the Poisson model. A shopping model example suggests the latter approach to be preferable.