Plot screening performance and unresolved outcomes

Usage

# S3 method for class 'mfrm_screening_performance'
plot(
  x,
  scope = c("family", "target"),
  metric = c("false_flags", "detection"),
  condition = NULL,
  draw = TRUE,
  ...
)

Arguments

x: Output from mfrm_screening_performance().
scope: "family" (default) shows any-flag events once per replication; "target" shows each individual target separately.
metric: "false_flags" (default) or "detection".
condition: Optional condition labels to include, matched exactly.
draw: Logical; FALSE returns exact plot data without opening a device.
...: Unused.

Value

Invisibly, an mfrm_plot_data object. Use plot_data() for custom graphics; automatic as_ggplot() conversion is unavailable.

Details

Blue points and whiskers show conditional rates and their exact binomial Monte Carlo intervals. Gray segments bound the realized all-trial proportion by assigning unresolved outcomes negative or positive; these are not confidence intervals. A row with no known outcomes has no point estimate. Right-hand counts show available/planned events. No model is fitted and no threshold is selected by this plot.