Sweep candidate clustering diagnostic

ledgr_sweep_cluster() clusters retained completed-candidate return series with one deterministic hierarchical method. The effective independent trial count is the number of clusters at distance_threshold; it is evidence for selection-integrity diagnostics, not a selection rule.

Usage

ledgr_sweep_cluster(sweep, candidates = NULL, distance_threshold = 0.5)

# S3 method for class 'ledgr_sweep_cluster'
as_tibble(
  x,
  what = c("summary", "membership", "distances"),
  ...
)

# S3 method for class 'ledgr_sweep_cluster'
print(x, ...)

Arguments

sweep

A ledgr_sweep_results object with retained completed returns.

candidates

Optional character vector of candidate ids to include.

distance_threshold

Numeric scalar in [0, 2]. Distances are 1 - correlation over retained return columns.

x

A ledgr_sweep_cluster object.

what

Table to return: "summary", "membership", or "distances".

...

Additional arguments passed to tibble printing.

Value

A ledgr_sweep_cluster object with summary, membership, distances, and metadata.

Details

The diagnostic consumes ledgr_sweep_returns_panel(..., value = "returns", complete = TRUE). It does not inspect fills, positions, promotion records, or walk-forward folds, and it does not mutate sweep artifacts or identity.

See also

vignette("selection-integrity", package = "ledgr") or system.file("doc", "selection-integrity.html", package = "ledgr").

Examples

if (FALSE) { # \dontrun{
clusters <- ledgr_sweep_cluster(sweep)
as_tibble(clusters)
as_tibble(clusters, what = "membership")
} # }