Scripting functionalities to produce multivariate clusters and outliers maps.
Run in a Python shell the Ci_Mmi.py script by specifying as input a vector spatial layer (see e.g. sample data). The output layer can be styled in QGIS using the output_sample_style.qml styling file.
See: Oxoli, D, Sabri, S, Rajabifard, A, Brovelli, MA. A classification technique for local multivariate clusters and outliers of spatial association. Transactions in GIS. 2020; 24: 1227– 1247. https://doi.org/10.1111/tgis.12639