Compute per-component variance explained — dkge_variance

Returns the standard deviation, variance, and cumulative variance explained by the DKGE components extracted in [dkge_fit()].

Usage

dkge_variance_explained(fit, relative_to = c("kept", "total"))

Arguments

fit: A `dkge` object.
relative_to: Compute variance proportions relative to "kept" (default, only the retained components) or "total" (all possible components).

Value

Data frame with columns `component`, `sdev`, `variance`, `prop_var`, and `cum_prop_var`.

Examples

toy <- dkge_sim_toy(
  factors = list(A = list(L = 2), B = list(L = 3)),
  active_terms = c("A", "B"), S = 3, P = 15, snr = 5
)
fit <- dkge(toy$B_list, toy$X_list, kernel = toy$K, rank = 2)
#> Warning: Argument 'kernel' is deprecated; use 'K' instead.
dkge_variance_explained(fit)
#>   component      sdev  variance  prop_var cum_prop_var
#> 1         1 11.914935 141.96569 0.5885228    0.5885228
#> 2         2  9.962835  99.25808 0.4114772    1.0000000