A Spatial Analysis of Primary Health Care Markets in Rural Areas
使用空间交互模型分析北加州农村居民就医选择模式，发现普通最小二乘法预测效果优于其他方法，并探讨了移除当地医生对患者流向的影响以及新医生可预期的市场份额。
Abstract An exponential spatial interaction model was used to analyze primary‐health‐care‐seeking patterns of rural consumers in northern California. A macrotheory of spatial interaction and a microtheory of consumer search were combined as a logically consistent rationale for the model. Ordinary least squares proved superior to nonlinear least squares and a limited dependent variable maximum likelihood technique in predicting health‐care‐seeking patterns. The model was used to address policy questions including the impact of removing local providers on patient flows to remaining providers and the market shares that could be expected by a new provider.