Formula interfaces for reference penalized GWR models — gwr_reference

These are concise formula wrappers around [gwr_penalized()].

Usage

gwr_lasso(formula, data, coords, ...)

gwr_ridge(formula, data, coords, ...)

gwr_en(formula, data, coords, ..., alpha = 0.5)

Arguments

formula: Model formula with an intercept.
data: Data frame containing model variables.
coords: Optional coordinate matrix used to construct neighbors.
...: Additional arguments passed to [gwr_penalized()].
alpha: Elastic-net mixing parameter in `[0, 1]`.

Value

An object inheriting from `gwrs_fit`.

Examples

quake <- datasets::quakes[1:30, ]
control <- gwrs_control(n_threads = 1)
lasso <- gwr_lasso(
  mag ~ depth + stations, quake, c("long", "lat"),
  k = 18, lambda = 0.05, control = control
)
ridge <- gwr_ridge(
  mag ~ depth + stations, quake, c("long", "lat"),
  k = 18, lambda = 0.05, control = control
)
elastic_net <- gwr_en(
  mag ~ depth + stations, quake, c("long", "lat"),
  k = 18, lambda = 0.05, alpha = 0.6, control = control
)
c(lasso$method, ridge$method, elastic_net$method)