Welcome to this Lattice Data with R Assignment!
In this exercise, we will explore the concepts and applications of spatial autocorrelation and extend our understanding of Ordinary Least Squares Regression (OLS) with Geographically Weighted Regression. We investigate:
1) Spatial Autocorrelation
a. Neighbors and Weight Matrices
b. Moran’s I, along with its Monte-Carlo derivative, test for spatial autocorrelation
c. Moran’s scatter plot and correlogram
d. Local Indicators of Spatial Autocorrelation (LISA)
3) Ordinary Least Squares (OLS) Regression and Geographically Weighted Regresssion (GWR)
4) we also briefly highlight ways we can interrogate the quality of a GWR with;
a. 3 spgwr significance tests
For this assignment we use a dataset that is well-suited to illustrate these concepts. Lattice data consisting of South African Police Service (crime data), City of Cape Town property valuation data and two variables from Statistics South Africa (STATSSA) census data. The dataset is discussed in the latticeData.ipynb.