Create a kernel weights by specifying k-nearest neighbors and a kernel method
kernel_knn_weights(
sf_obj,
k,
kernel_method,
adaptive_bandwidth = TRUE,
use_kernel_diagonals = FALSE,
power = 1,
is_inverse = FALSE,
is_arc = FALSE,
is_mile = TRUE
)An sf (simple feature) object
a positive integer number for k-nearest neighbors
a string value, which has to be one of 'triangular', 'uniform', 'epanechnikov', 'quartic', 'gaussian'
(optional) TRUE (default) or FALSE: TRUE use adaptive bandwidth calculated using distance of k-nearest neithbors, FALSE use max distance of all observation to their k-nearest neighbors
(optional) FALSE (default) or TRUE, apply kernel on the diagonal of weights matrix
(optional) The power (or exponent) of a number says how many times to use the number in a multiplication.
(optional) FALSE (default) or TRUE, apply inverse on distance value
(optional) FALSE (default) or TRUE, compute arc distance between two observations
(optional) TRUE (default) or FALSE, convert distance unit from mile to km.
An instance of Weight-class
library(sf)
guerry_path <- system.file("extdata", "Guerry.shp", package = "rgeoda")
guerry <- st_read(guerry_path)
#> Reading layer `Guerry' from data source
#> `/Users/runner/work/_temp/Library/rgeoda/extdata/Guerry.shp'
#> using driver `ESRI Shapefile'
#> Simple feature collection with 85 features and 29 fields
#> Geometry type: MULTIPOLYGON
#> Dimension: XY
#> Bounding box: xmin: 47680 ymin: 1703258 xmax: 1031401 ymax: 2677441
#> Projected CRS: NTF (Paris) / Lambert zone II
adptkernel_w = kernel_knn_weights(guerry, 6, "uniform")
summary(adptkernel_w)
#> name value
#> 1 number of observations: 85
#> 2 is symmetric: FALSE
#> 3 sparsity: 0.0705882352941176
#> 4 # min neighbors: 6
#> 5 # max neighbors: 6
#> 6 # mean neighbors: 6
#> 7 # median neighbors: 6
#> 8 has isolates: FALSE