Heavyweight spatial statistics for TPS data

R parity of morie.tps_spatial_advanced. Builds on tps_spatial (global Moran's I, LISA, KDE) with:

Details

• morie_tps_ripley_k: Ripley's K function for point-pattern clustering at multiple radii.

• morie_tps_getis_ord_g_star: local Getis-Ord Gi* hot/cold-spot z-scores.

• morie_tps_dbscan_clusters: density-based clusters on lat/long (via dbscan, optional).

• morie_tps_polygon_morans_i: polygon-aware Moran's I from an sf object's actual polygon centroids (instead of the centroid-only k-NN approximation in morie_tps_morans_i_neighbourhood).

• morie_tps_bivariate_moran: bivariate Moran's I between two attributes at the same polygons.

• morie_tps_moran_sweep_heatmap: a (category x year) sweep of polygon Moran's I.

Polygon functions accept either an sf object (gated with requireNamespace("sf")) or a plain data.frame carrying precomputed centroid columns. KNN graphs prefer FNN; DBSCAN requires the optional dbscan package; spatial autocorrelation tests can optionally be delegated to spdep.