Figure 5 plotting function Lambda uncertainty plot (per player)

Forest plot of \(\lambda\) with uncertainty intervals, using relabeled draws. Points colored by the hard partition (fit$estimates$x_hat).

Usage

plot_lambda_uncertainty(
  fit,
  w_ij = NULL,
  players_df = NULL,
  players_id_col = NULL,
  labels = NULL,
  labels_key_col = NULL,
  labels_value_col = NULL,
  clean_display_labels = NULL,
  order_ids = NULL,
  log_base = exp(1),
  max_n_clust = NULL,
  prob = 0.9,
  palette = NULL,
  clean_fun = function(x) x,
  x_hat = NULL,
  filter_lambdas = TRUE,
  conditional = TRUE,
  ...
)

Arguments

fit

Output from gibbs_BT_SBM() with opt_lambda$lambda_item computed (set keep_lambda=TRUE when sampling).

w_ij

Optional wins matrix to compute marginal wins for ordering and annotation.

players_df

Optional data.frame used to infer display labels.

players_id_col

Optional id column name in players_df.

labels

Optional display labels; either named by item ids or aligned with items.

labels_key_col

Optional key column in players_df used to match item ids.

labels_value_col

Optional value column in players_df containing display labels.

clean_display_labels

Logical; if TRUE, applies clean_fun to resolved display labels.

order_ids

Optional explicit player ordering.

log_base

Base for the x-axis logarithm (10 or e). Defaults to 10.

max_n_clust

Where to filter the mcmc x_t. If not specified we use the modal K

prob

Interval probability for HPD (e.g., 0.90).

palette

Named colors for clusters.

clean_fun

Optional player-name cleaner. Default: identity.

x_hat

Optional hard partition; if NULL, inferred from fit.

filter_lambdas

Logical; when conditional = TRUE, keeps lambda draws aligned to filtered x draws.

conditional

Logical; if TRUE, intervals are computed conditional on each item's hard cluster.

...

Reserved for future extensions.

Value

A ggplot object.

Examples

if (FALSE) { # \dontrun{
# fit with keep_lambda=TRUE
p <- plot_lambda_uncertainty(fit, prob = 0.90)
} # }