GeoDevelopment Observatory of Bolivia / Tools
Spatial distribution and dependence of poverty
Two ways to classify a poverty map, then spatial weights, the spatial lag, Moran’s I and local clusters.
This Colab notebook first maps the Multidimensional Poverty Index of 2012 with box-plot breaks, which flag three high outliers (El Sena, San Pedro and Poroma), and with three Fisher–Jenks natural breaks. It then builds six-nearest-neighbour weights, shows that Moran’s I (0.46) is the slope of the spatial-lag regression, and maps local clusters and outliers, comparing the 10% and 5% significance levels and how many clusters would be expected by chance.