This chapter deals with

• Spatial autocorrelation and its importance to geographical problems

• Global and local spatial autocorrelation techniques like Moran’s I, Getis-Ord G and Geary C

• Tracing spatial clusters of high values (hot spots) or low values (cold spots)

• Tracing spatial outliers

• Optimized hot spot analysis

• Interpreting the statistical significance of results

• Incremental spatial autocorrelation used to define the appropriate scale of analysis

• The multiple comparison problem and spatial dependence

• Introducing Bonferroni correction and the false discovery rate

• Spatiotemporal autocorrelation analysis using bivariate and differential Local Moran’s I index

• Presenting examples of step-by-step solutions using ArcGIS and GeoDa

After a thorough study of the theory and lab sections, you will be able to

• Distinguish between global and local spatial autocorrelation

• Understand why spatial autocorrelation analysis is relevant to geographical analysis

• Apply local and global indices of spatial autocorrelation like local Moran’s, Getis-Ord Gi and Gi∗

• Use Moran’s I scatter plot to identify patterns

• Identify if clustering of hot or cold spots exist

• Identify and locate spatial outliers

• Use Bivariate and Differential Local Moran’s I to identify if spatiotemporal autocorrelation exists and if changes cluster over time

• Apply these tools using ArcGISand GeoDa

• Interpret the results from both the statistical significance and spatial analysis standpoints