Optimize Spatial Block Size — borg_block

Tests multiple block sizes and selects the one that minimizes residual spatial autocorrelation between CV folds. For each candidate size, generates spatial block folds and computes the mean Moran's I of residuals within test sets.

Usage

borg_block_size(
  data,
  coords,
  target,
  v = 5,
  n_sizes = 10,
  range = NULL,
  formula = NULL,
  crs = NULL,
  verbose = FALSE
)

Arguments

data: Data frame with coordinate and target columns.
coords: Character vector of length 2. Coordinate column names.
target: Character. Target variable name.
v: Integer. Number of folds. Default: 5.
n_sizes: Integer. Number of candidate block sizes to test. Default: 10.
range: Numeric vector of length 2. Min and max block sizes to test. If NULL, automatically determined from the variogram range estimate (0.5x to 3x range).
formula: Optional model formula for computing residual autocorrelation. If NULL, uses the raw target values.
crs: Optional coordinate reference system (an sf CRS, EPSG code, or PROJ string). When it is geographic (lon/lat), distances use the Haversine formula; otherwise planar Euclidean.
verbose: Logical. Default: FALSE.

Value

A list with class "borg_block_opt" containing:

optimal: Optimal block size
results: Data frame with columns: block_size, mean_morans_i, mean_test_size, n_empty_folds
range_estimate: Variogram-based range estimate

Has an autoplot() method showing the optimization curve.

Examples

set.seed(42)
d <- data.frame(
  x = runif(200, 0, 100), y = runif(200, 0, 100),
  z = rnorm(200)
)
opt <- borg_block_size(d, coords = c("x", "y"), target = "z")
opt$optimal