Runs adapter-owned `multifer` component-significance tests for either the COVSTATIS interstructure/RV axes across input tables or the compromise ROI eigencomponents.
Usage
infer_covstatis(
x,
data,
axes = c("compromise", "interstructure", "both"),
B = 1000L,
alpha = 0.05,
seed = NULL,
strict = TRUE,
...
)Arguments
- x
A fitted `covstatis` object.
- data
Original list of covariance/correlation matrices used to fit `x`.
- axes
Which component family to test: `"interstructure"` tests axes of the table-by-table RV/interstructure matrix, `"compromise"` tests ROI eigencomponents of the weighted compromise matrix, and `"both"` runs both tests and returns a named list.
- B
Number of Monte Carlo draws per ladder rung.
- alpha
Significance threshold passed to [multifer::infer()].
- seed
Optional random seed.
- strict
Logical; passed to [multifer::infer()].
- ...
Additional arguments passed to [multifer::infer()].