Spatial Prediction Uncertainty Map — borg_prediction

Collects predictions from all CV folds at each test location, computes per-location mean and standard deviation, and provides a spatial uncertainty map.

Usage

borg_prediction_map(data, folds, formula, coords, fit_fun = stats::lm)

Arguments

data: Data frame with predictor and coordinate columns.
folds: A borg_cv object or fold list.
formula: Model formula.
coords: Character vector of length 2. Coordinate column names.
fit_fun: Function. Default: lm.

Value

A data frame with class "borg_pred_map" containing:

x, y: Coordinates
pred_mean: Mean prediction across folds where this obs was in test set
pred_sd: SD of predictions (uncertainty)
actual: Observed target value
residual_mean: Mean residual
n_folds_tested: Number of folds where this obs appeared in test set

Examples

set.seed(42)
d <- data.frame(x = runif(100), y = runif(100), z = rnorm(100))
cv <- borg_cv(d, coords = c("x", "y"), target = "z")
pm <- borg_prediction_map(d, cv, z ~ x + y, coords = c("x", "y"))
head(pm)