Produces a compact summary of the three most statistically significant transfer entropy estimates across directions and lags, optionally combining information from counts, rates, and binary specifications.
summarise_te_top3_posthoc(te_tab, dir_csv)A data frame with up to three rows and columns:
model: constant string "TransferEntropy".
config: textual description of direction, lag, and,
if available, type (counts, rates, binary).
stat: transfer entropy estimate.
p_value: associated p-value.
If no results are available, a single-row data frame with NA
entries is returned.
Optional data frame with transfer entropy results,
containing at least columns lag, TE_ItoC,
TE_CtoI, p_ItoC, p_CtoI, and optionally
type. If NULL or empty, the function attempts to read
the data from CSV files via the internal helper .read_te_all().
Character scalar; directory where the transfer entropy
CSV files are stored (used when te_tab is missing).
The function reshapes te_tab into a long format with directions
"I->C" and "C->I", orders by p-value (ascending) and
lag, and keeps the three rows with the smallest p-values.