robust.percapita: Robust per capita rates from gridded data

In rafaelgramos/robustmap: Robust mapping of spatiotemporal data

Robust per capita rates from gridded data

Description

Estimate robust per capita rates from a event count map (numerator) and a reference population map (denominator); both need to be SpatRaster grids with the same dimensions and covering the same extent. Event count map can be generated using robust.grid or robust.quadcount from a point set.

Usage

robust.percapita(numerator, denominator, bd = NULL, weighted = FALSE)

Arguments

numerator	an event count grid (SpatRaster)
denominator	a reference population map grid (SpatRaster), of the same dimensions, extent and resolution as numerator.
bd	bandwidth to be used by the GWR model. If NULL (default), bandwidth is estimated by cross-validation (using spgwr::gwr.sel())
weighted	if true, denominator is used also as weights for the residuals in a Weighted Least Squares fashion. The rational is that areas with a greater denominator (e.g. more individuals experiencing a per capita rate) should have greater weight in the estimation of the rate.

Author(s)

Rafael G. Ramos (main proponent and coder)

References

Ramos, R. G. (2021). Improving victimization risk estimation: A geographically weighted regression approach. ISPRS International Journal of Geo-Information, 10(6), 364.

rafaelgramos/robustmap documentation built on April 22, 2024, 8:22 a.m.