Abstract
Negative spatial autocorrelation is one of the most neglected concepts in quantitative geography, regional science, and spatial statistics/econometrics in general. This paper focuses on and contributes to the literature in terms of the following three reasons why this neglect exists: Existing spatial autocorrelation quantification, the popular form of georeferenced variables studied, and the presence of both hidden negative spatial autocorrelation, and mixtures of positive and negative spatial autocorrelation in georeferenced variables. This paper also presents details and insights by furnishing concrete empirical examples of negative spatial autocorrelation. These examples include: Multi-locational chain store market areas, the shrinking city of Detroit, Dallas-Fort Worth journey-to-work flows, and county crime data. This paper concludes by enumerating a number of future research topics that would help increase the literature profile of negative spatial autocorrelation.
Highlights
Quantitative data analysis researchers routinely initiate their studies by examining univariate features of their data, such as frequency distribution skewness, the relationship between a mean and its accompanying variance, and the presence of outliers
If the trend line representing scatter of polygons points is zeroclosest lengthtoboundaries
Griffith and Chun [20] furnish one of a number of available overviews of Moran Eigenvector Spatial Filtering (MESF), an innovative spatial statistical methodology that is an alternative to spatial autoregression, and that adds a set of synthetic proxy variables—eigenvectors extracted from matrix expression (2), an adjusted spatial weights matrix that describes connections among geographic objects in space—as control variables to filter spatial autocorrelation out of regression residuals and transfer it to the mean response in a model specification; the intercept term changes from a constant to a variable, like with mixed model specifications
Summary
Quantitative data analysis researchers routinely initiate their studies by examining univariate features of their data, such as frequency distribution skewness, the relationship between a mean and its accompanying variance, and the presence of outliers (e.g., see reference [1]) If these researchers study two or more variables, they habitually inspect pairwise linear correlation coefficients. These latter include autocorrelation, literally self-correlation, toto correlation These latter include autocorrelation,which which literallymeans means self-correlation,and andrefers refers correlation among observation a single variable, a fundamental topic ofof spatial statistics. A unit of analysis variable value pairings of traditional linear correlation are by observation:. The measurement value pairings of spatial autocorrelation pertain to datapertain organization variable attributes. A perusal of relevant textbooks reveals the presentation ofreveals very few world NSAofexamples This concept the spatial topics, and asspatial such isstatistics the motivation for and theme of this article.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call