PRIOR STRUCTURE AND SPATIAL INTERACTION

A rationale for continued interest in simple, static urban models, such as the gravity-type models, is given through the perspective of the notion of prior structure, according to which there is, embedded in every model, a largely non-empirical component resulting from the structural properties of the theoretical language used. It is suggested that the prior structure implicit in gravity models is responsible for the widely observed map effect in spatial interaction as well as for the relative predictive power of these models. A procedure is outlined for deriving that prior structure from first principles. Combined with a minimal amount of behavioural information, that quasi-analytic component can be used as prior information, in the Bayesian sense, to generate familiar forms of the gravity model as the most probable consequences of svstem specification. The practical significance of the prior-structure effect for model performance is then briefly discussed.

[1]  Chang-i Hua,et al.  A Critical Review of the Development of the Gravity Model , 1979 .

[2]  S. Stein,et al.  Calculus and Analytic Geometry , 1973 .

[3]  Michael Batty,et al.  Generalized Measures of Information, Bayes' Likelihood Ratio and Jaynes' Formalism , 1975 .

[4]  Waldo R. Tobler,et al.  Spatial Interaction Patterns , 1976 .

[5]  I. Good,et al.  The Maximum Entropy Formalism. , 1979 .

[6]  L. Curry A Spatial Analysis of Gravity Flows , 1972, The Random Spatial Economy and Its Evolution.

[7]  H. Williams On the Formation of Travel Demand Models and Economic Evaluation Measures of User Benefit , 1977 .

[8]  Howard Hunt Pattee,et al.  The complementarity principle in biological and social structures , 1978 .

[9]  G. Nerlich The shape of space , 1976 .

[10]  P. Gould Is Statistix Inferens the Geographical Name for A Wild Goose , 1970 .

[11]  W. Tobler A Model of Geographical Movement , 1981 .

[12]  Helen Couclelis,et al.  The Notion of Prior Structure in Urban Modelling , 1984 .

[13]  A. Anas Discrete choice theory, information theory and the multinomial logit and gravity models , 1983 .

[14]  E. Jaynes Information Theory and Statistical Mechanics , 1957 .

[15]  E. Jaynes Foundations of Probability Theory and Statistical Mechanics , 1967 .

[16]  T E Smith,et al.  A Cost-Efficiency Approach to the Analysis of Congested Spatial-Interaction Behavior , 1983 .

[17]  P. Gould ACQUIRING SPATIAL INFORMATION , 1975 .

[18]  H. Timmermans,et al.  Behavioural Models and Spatial Planning: Some Methodological Considerations and Empirical Tests , 1981 .

[19]  Tony E. Smith,et al.  A choice theory of spatial interaction , 1975 .

[20]  W. Tobler,et al.  An Alternative Formulation for Spatial-Interaction Modeling , 1983 .

[21]  Alan Wilson,et al.  Entropy in urban and regional modelling , 1972, Handbook on Entropy, Complexity and Spatial Dynamics.

[22]  Michael Batty,et al.  The Method of Residues in Urban Modelling , 1976 .

[23]  Eric Sheppard,et al.  NOTES ON SPATIAL INTERACTION , 1979 .

[24]  A. Fotheringham SPATIAL STRUCTURE AND DISTANCE‐DECAY PARAMETERS , 1981, Annals of the Association of American Geographers.

[25]  M. Tribus Rational descriptions, decisions, and designs , 1969 .

[26]  Eric Sheppard THE IDEOLOGY OF SPATIAL CHOICE , 1980 .