Skip to contents

Perform 1D density estimation, compute and plot the resulting highest density regions in a way similar to ggplot2::geom_rug(). Note, the plotted objects have the probs mapped to the alpha aesthetic by default.


  mapping = NULL,
  data = NULL,
  geom = "hdr_rug",
  position = "identity",
  method = "kde",
  probs = c(0.99, 0.95, 0.8, 0.5),
  xlim = NULL,
  ylim = NULL,
  h = "nrd0",
  adjust = 1,
  kernel = "gaussian",
  bins = NULL,
  n = 512,
  na.rm = FALSE,
  show.legend = TRUE,
  inherit.aes = TRUE

  mapping = NULL,
  data = NULL,
  stat = "hdr_rug",
  position = "identity",
  outside = FALSE,
  sides = "bl",
  length = unit(0.03, "npc"),
  na.rm = FALSE,
  show.legend = TRUE,
  inherit.aes = TRUE



Set of aesthetic mappings created by aes() or aes_(). If specified and inherit.aes = TRUE (the default), it is combined with the default mapping at the top level of the plot. You must supply mapping if there is no plot mapping.


The data to be displayed in this layer. There are three options:

If NULL, the default, the data is inherited from the plot data as specified in the call to ggplot().

A data.frame, or other object, will override the plot data. All objects will be fortified to produce a data frame. See fortify() for which variables will be created.

A function will be called with a single argument, the plot data. The return value must be a data.frame, and will be used as the layer data. A function can be created from a formula (e.g. ~ head(.x, 10)).


The geometric object to use display the data


Position adjustment, either as a string, or the result of a call to a position adjustment function.


Other arguments passed on to layer(). These are often aesthetics, used to set an aesthetic to a fixed value, like colour = "red" or size = 3. They may also be parameters to the paired geom/stat.


Density estimator to use, accepts character vector: "kde", "histogram", "freqpoly", or "mvnorm".


Probabilities to compute highest density regions for.

xlim, ylim

Range to compute and draw regions. If NULL, defaults to range of data.


The smoothing bandwidth to be used. If numeric, the standard deviation of the smoothing kernel. If character, a rule to choose the bandwidth, as listed in stats::bw.nrd().


A multiplicate bandwidth adjustment. This makes it possible to adjust the bandwidth while still using the a bandwidth estimator. For example, adjust = 1/2 means use half of the default bandwidth.


Kernel. See list of available kernels in density().


Number of bins along each axis for histogram and frequency polygon estimators. Either a vector of length 2 or a scalar value which is recycled for both dimensions. Defaults to normal reference rule (Scott, pg 87).


Resolution of grid used in discrete approximations for kernel density and parametric estimators.


If FALSE, the default, missing values are removed with a warning. If TRUE, missing values are silently removed.


logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped. FALSE never includes, and TRUE always includes. It can also be a named logical vector to finely select the aesthetics to display.


If FALSE, overrides the default aesthetics, rather than combining with them. This is most useful for helper functions that define both data and aesthetics and shouldn't inherit behaviour from the default plot specification, e.g. borders().


The statistical transformation to use on the data for this layer, as a string.


logical that controls whether to move the rug tassels outside of the plot area. Default is off (FALSE). You will also need to use coord_cartesian(clip = "off"). When set to TRUE, also consider changing the sides argument to "tr". See examples.


A string that controls which sides of the plot the rugs appear on. It can be set to a string containing any of "trbl", for top, right, bottom, and left.


A grid::unit() object that sets the length of the rug lines. Use scale expansion to avoid overplotting of data.


geom_hdr_rug understands the following aesthetics (required aesthetics are in bold):

  • alpha

  • fill

  • group

  • subgroup

  • x

  • y

Computed variables


The probability of the highest density region, specified by probs, corresponding to each point.


Scott, David W. Multivariate Density Estimation (2e), Wiley.


df <- data.frame(x = rnorm(100), y = rnorm(100))

# Plot marginal HDRs for bivariate data
ggplot(df, aes(x, y)) +
  geom_point() +
  geom_hdr_rug() +

ggplot(df, aes(x, y)) +
  geom_hdr() +
  geom_hdr_rug() +

# Or, plot marginal HDR for univariate data
ggplot(df, aes(x)) +
  geom_density() +

ggplot(df, aes(y = y)) +
  geom_density() +

# Can specify location of marginal HDRs as in ggplot2::geom_rug(),
ggplot(df, aes(x, y)) +
  geom_hdr() +
  geom_hdr_rug(sides = "tr", outside = TRUE) +
  coord_fixed(clip = "off")

# Can use same methods of density estimation as geom_hdr().
# For data with constrained support, we suggest setting method = "histogram":
ggplot(df, aes(x^2)) +
 geom_histogram(bins = 30, boundary = 0) +
 geom_hdr_rug(method = "histogram")

ggplot(df, aes(x^2, y^2)) +
 geom_hdr(method = "histogram") +
 geom_hdr_rug(method = "histogram") +