Parameter Estimation in Spatial Interaction Modeling

1973 ◽  
Vol 5 (4) ◽  
pp. 503-518 ◽  
Author(s):  
F J Cesario

New ways of estimating the parameters of the so-called ‘doubly-constrained’ spatial interaction model are developed and tested against contemporary procedures. Two variations of the doubly-constrained model—growth-factor and interactance (that is, gravity and entropy) models—are considered: The structural characteristics of these models are specified and several new solution properties are established. Then alternative parameter-estimation strategies are formulated and tested. The proposed techniques (which solve problems that are equivalent to existing models) are shown to be inferior to current methodology with respect to all criteria related to solution speed. However, it is shown that the proposed methods are more flexible than current approaches since a wider range of assumptions can be incorporated into the analysis.

1983 ◽  
Vol 15 (5) ◽  
pp. 705-706 ◽  
Author(s):  
A G Wilson

It is shown that there is a possible relationship between the maximand used to produce To bier's spatial interaction model and Kullback entropy.


1992 ◽  
Vol 24 (3) ◽  
pp. 441-446 ◽  
Author(s):  
P B Slater

Spatial interaction modeling can be embedded in the framework of statistical thermodynamics. This has been shown in part by Wilson and by Erlander, in particular. However, to complete the embedding, it is necessary to ask the dynamic question of how a migratory population evolves over time from one (observed) distribution at time t0 to another (observed) distribution at time t1, rather than just the question of how many people who started in one area at t0 ended in another at t1—with no regard to the intermediate locations of the migrants. The manner in which such evolutionary processes are addressed in the context of equilibrium and nonequilibrium thermodynamics serves as a basis for extending the spatial interaction model of human movement.


1979 ◽  
Vol 11 (5) ◽  
pp. 527-539 ◽  
Author(s):  
Irena Chudzyńska ◽  
Z Słodkowski

A mathematical model of urban spatial interaction based on the intervening-opportunities principle is discussed and its equilibria are studied. It is shown that, under natural assumptions, the number of equilibria is finite, and a mathematical criterion for distinguishing the equilibrium corresponding to reality is given.


REGION ◽  
2016 ◽  
Vol 3 (2) ◽  
pp. 11 ◽  
Author(s):  
Taylor M Oshan

This primer provides a practical guide to get started with spatial interaction modeling using the SpInt module in the python spatial analysis library (PySAL).


Sign in / Sign up

Export Citation Format

Share Document