I am using the package GWmodel to predict at a finer spatial scale.
Basically, I want my output to be a single raster layer. For the GWR, I am
using one dependent and 2 independent variables. I don't know how to create
the prediction raster.
When using only one independent variable it's easy. For example:
library(GWmodel)
library(sp)
library(raster)
wd = "path/"
provoliko = "EPSG:7767"
# this is my fine res raster
tirs = raster(paste0(wd, "tirs.tif"))
regpoints <- as(tirs, "SpatialPoints")
# this df contains the coarse res variables
block.data = read.csv(paste0(wd, "block.data.csv"))
coordinates(block.data) <- c("x", "y")
proj4string(block.data) <- provoliko
eq1 <- ntl ~ tirs
abw = bw.gwr(eq1,
data = block.data,
approach = "AIC",
kernel = "gaussian",
adaptive = TRUE,
p = 2,
parallel.method = "omp",
parallel.arg = "omp")
ab_gwr = gwr.basic(eq1,
data = block.data,
regression.points = regpoints,
bw = abw,
kernel = "gaussian",
adaptive = TRUE,
p = 2,
F123.test = FALSE,
cv = FALSE,
parallel.method = "omp",
parallel.arg = "omp")
ab_gwr
sp <- ab_gwr$SDF
sf <- st_as_sf(sp)
# intercept
intercept = as.data.frame(sf$Intercept)
intercept = SpatialPointsDataFrame(data = intercept, coords = regpoints)
gridded(intercept) <- TRUE
intercept <- raster(intercept)
raster::crs(intercept) <- provoliko
# slope
slope = as.data.frame(sf$tirs)
slope = SpatialPointsDataFrame(data = slope, coords = regpoints)
gridded(slope) <- TRUE
slope <- raster(slope)
raster::crs(slope) <- provoliko
gwr_pred = intercept + slope * s
writeRaster(gwr_pred,
paste0(wd, "ntl_gwr.tif"),
overwrite = TRUE)
When using multiple independent variables, I tried to stack them before I
convert them to spatialPoints, but then I don't know what to set in the
slope parameter.
--
Tziokas Nikolaos
Cartographer
Tel:(+44)07561120302
LinkedIn <http://linkedin.com/in/nikolaos-tziokas-896081130>
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
