Skip to contents

Significance region geom

Source: R/geoms-regions.R
geom_cpt_region.Rd

Draws a changepoint that is an interval: a vertical band spanning xmin to xmax and the full height of the panel. This is the display Narrowest Significance Pursuit needs (nsp_wrapper()): NSP returns intervals each of which contains at least one changepoint at a prescribed global significance level, which is neither a point estimate nor a confidence interval around one. autoplot() adds this layer automatically for a result that carries regions.

Usage

geom_cpt_region(
  mapping = NULL,
  data = NULL,
  ...,
  alpha = 0.2,
  fill = "steelblue",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = FALSE
)

Arguments

mapping

Aesthetic mappings. Requires xmin and xmax; ymin/ymax default to the panel extent, so they need not be supplied.

data

A data frame of regions, e.g. cpt_regions() output.

...

Other arguments passed to ggplot2::geom_rect().

alpha

Fill transparency. Defaults to 0.2 — light enough that the series stays readable through overlapping bands.

fill

Band fill colour. Defaults to "steelblue".

na.rm

If FALSE, missing values are removed with a warning.

show.legend

Whether to show a legend.

inherit.aes

Whether to inherit the plot's aesthetics. Defaults to FALSE: a region frame has its own columns and nothing to do with the series' x/y.

Value

A ggplot layer.

See also

nsp_wrapper(), cpt_regions().

Other ggplot2 layers: geom_changepoint(), geom_cpt_ci(), geom_cpt_event(), geom_cpt_label(), geom_cpt_segment(), stat_changepoint()

Examples

library(ggplot2)
set.seed(2026)
d <- data.frame(t = 1:100, y = c(rnorm(50), rnorm(50, 4)))
regions <- data.frame(xmin = 44, xmax = 57)
ggplot(d, aes(t, y)) +
  geom_cpt_region(aes(xmin = xmin, xmax = xmax), data = regions) +
  geom_line()