The Geographic Diversity of U.S. Nonmetropolitan Growth Dynamics: A Geographically Weighted Regression Approach
研究使用地理加权回归分析美国非都市县就业增长的空间异质性，发现传统全局回归掩盖了地方差异，尤其是便利设施和大学毕业生比例的影响在不同地区显著不同。
<i>Spatial heterogeneity is introduced as an explanation for local-area growth mechanisms, especially employment growth. As these effects are difficult to detect using conventional regression approaches, we use Geographically Weighted Regressions (GWR) for non-metropolitan U.S. counties. We test for geographic heterogeneity in the growth parameters and compare them to global regression estimates. The results indicate significant heterogeneity in the regression coefficients across the country, most notably for amenities and college graduate shares. Using GWR also exposes significant local variations that are masked by global estimates suggesting limitations of a one-size-fits-all approach to describe growth and to inform public policy</i>.