A New Set of Spatial-Interaction Models: The Theory of Competing Destinations †
A New Set of Spatial-Interaction Models: The Theory of Competing Destinations †
复制标题
DOI：
10.1177/0308518x8301500103
复制
发表时间：
1983-01
期刊：
影响因子：
--
通讯作者：
A. Fotheringham
中科院分区：
文献类型：
--
作者：
A. Fotheringham
通常被称为重力模型的空间相互作用模型家族的成员被证明是错误指定的。这一错误规定的一个结果是在估计的距离衰减参数中出现了不希望看到的“空间结构效应”，并对这种影响进行了详细的研究。建立了一组可供选择的空间相互作用模型，从中可以获得更准确的相互作用预测和更准确的参数估计。这些新的相互作用模型被称为竞争目的地模型，在它们的校准中获得的估计距离衰减参数被证明具有纯粹的行为解释。文中还讨论了重力模型错误描述的影响。
Members of the family of spatial-interaction models commonly referred to as gravity models are shown to be misspecified. One result of this misspecification is the occurrence of an undesirable ‘spatial-structure effect’ in estimated distance-decay parameters and this effect is examined in detail. An alternative set of spatial-interaction models is formulated from which more accurate predictions of interactions and more accurate parameter estimates can be obtained. These new interaction models are termed competing destinations models, and estimated distance-decay parameters obtained in their calibration are shown to have a purely behavioural interpretation. The implications of gravity-model misspecification are discussed.