Given the candidate subsets and the indicators of the acceptable family of subsets, compute for each variable the proportion of acceptable subsets in which that variable appears. If specified, variable co-appearances can be computed as reported as well.
Arguments
- indicators
L x pmatrix of inclusion indicators (booleans) where each row denotes a candidate subset- all_accept
indices (i.e., rows of
indicators) that correspond to the acceptable subsets- co
logical; if TRUE, compute and return the co-variable importances
- xnames
the names of the x-variables