Given the output of select_optimal_K_r_mfa, this function computes
additional metrics for each fitted model in all_models, such as
global VAF (compute_global_vaf_mfa) and cluster sizes
(compute_cluster_sizes).
Arguments
- list_of_data
The same list of data used to fit the models.
- selection_obj
The object returned by
select_optimal_K_r_mfa.
Value
A data frame with columns:
- K
Number of clusters.
- r
Factor dimension.
- logLik
Log-likelihood from the fit.
- BIC
BIC from the fit.
- GlobalVAF
Global variance accounted for, \(1 - \mathrm{SSE}/\mathrm{SST}\).
- minClusterSize
Minimum cluster size.
- maxClusterSize
Maximum cluster size.
Details
This function loops over each entry in selection_obj$all_models,
extracts the model, calls compute_global_vaf_mfa and
compute_cluster_sizes, and records the results in a combined table.
Examples
if (FALSE) { # \dontrun{
sel_obj <- select_optimal_K_r_mfa(my_data, Kvec=2:4, rvec=1:3)
df_posthoc <- posthoc_mfa_evaluation(my_data, sel_obj)
head(df_posthoc)
} # }