Stein-Freiler Distinguished Service Professor of Sociology and the College at the University of Chicago
Chair, Committee on Geographical Sciences
Director, Center for Spatial Data Science
Senior Fellow, NORC
More LISA, Multivariate Extensions
It has been more than twenty years since the publication of the article on Local Indicators of Spatial Association (LISA) in Geographical Analysis (Anselin 1995). This article turned out to be one of the most cited and downloaded pieces in GA. Since its publication, the idea of a local indicator has led to a wide range of applications to spatial cluster and outlier detection, as well as to a considerable methodological literature dealing with extensions to different types of data (categorical, space-time), proper inference (the multiple comparison problem), and computational issues. In this presentation, I start with a brief review of these developments, with a focus on inference and computational issues. I then present some new results on a particular variant of LISA, the Local Geary statistic. I also generalize the univariate case to a Multivariate Indicator of Local Spatial Autocorrelation that takes the form of a multivariate Local Geary statistic. I discuss the statistical properties, inference, visualization and interpretation, and illustrate this with an example using Guerry’s 1833 classic data set on moral statistics in France.
Friday, April 13
10:30am -12pm CT
Social Science Research Center – Second Floor of McGiffert House (above Coop Bookstore / Plein Air)
5751 South Woodlawn Avenue
University of Chicago
Chicago, iL 60637
Multivariate Local Geary Map (Anselin, Luc. (2018). A Local Indicator of Multivariate Spatial Association: Extending Geary’s c. Geographical Analysis).