The Advanced Forum on“Spatial Data Analytics for Social Sciences”

Workshop on Recent Developments in Local Spatial Cluster Detection

Luc Anselin

Center for Spatial Data Science

University of Chicago

The concept of a Local Indicator of Spatial Association (LISA) has been around for more than 20 years and has seen many refinements since its initial introduction. In this workshop, the fundamental idea behind this statistic is reviewed and a number of recent univariate and multivariate extensions are outlined. The methods are illustrated through their implementation in the latest version of the open source GeoDa software.

Day 1: Dec 12, 2019

Univariate Analysis

• The many faces of spatial clusters
• LISA principle
• Local Moran and local Geary
• Getis-Ord statistics
• Local join count statistics
• Quantile LISA
• Interpretation

Day 2: Dec 13, 2019

Multivariate Analysis

• The multivariate problem
• Spatializing classic multidimensional analysis
• Multivariate local Geary
• Multivariate local co-location statistics
• Multivariate Quantile LISA
• Future directions