Skip to contents

Empirical semivariogram of residuals

Source: R/diagnostics_sim.R
tulpa_variogram.Rd

Computes the empirical semivariogram in distance bins for visual assessment of remaining spatial structure in residuals.

Usage

tulpa_variogram(
  object,
  coords,
  n_bins = 15L,
  max_dist = NULL,
  resid_type = "pearson"
)

# S3 method for class 'tulpa_variogram'
plot(x, ...)

# S3 method for class 'tulpa_variogram'
print(x, ...)

Arguments

object

A fitted model, or a numeric vector of residuals

coords

N x 2 coordinate matrix (required)

n_bins

Number of distance bins (default 15)

max_dist

Maximum distance (default: half the maximum pairwise distance)

resid_type

Residual type if extracting from model (default "pearson")

x

A tulpa_variogram table.

...

Passed to graphics::plot() by plot() and to print() by print().

Value

A tulpa_variogram data.frame with columns dist, gamma, n_pairs

plot() draws the semivariance against distance, point size scaled by the pair count, and returns x invisibly.

Examples

set.seed(1)
coords <- cbind(runif(60), runif(60))
tulpa_variogram(rnorm(60), coords, n_bins = 5)