Introduction to ggridges

Ridgelines

The geom geom_ridgeline can be used to draw lines with a filled area underneath.

library(ggplot2)
library(ggridges)

data <- data.frame(x = 1:5, y = rep(1, 5), height = c(0, 1, 3, 4, 2))
ggplot(data, aes(x, y, height = height)) + geom_ridgeline()

Negative heights are allowed, but are cut off unless the min_height parameter is set negative as well.

library(patchwork) # for side-by-side plotting

data <- data.frame(x = 1:5, y = rep(1, 5), height = c(0, 1, -1, 3, 2))
plot_base <- ggplot(data, aes(x, y, height = height))

plot_base + geom_ridgeline() | plot_base + geom_ridgeline(min_height = -2)

Multiple ridgelines can be drawn at the same time. They will be ordered such that the ones drawn higher up are in the background. When drawing multiple ridgelines at once, the group aesthetic must be specified so that the geom knows which parts of the data belong to which ridgeline.

d <- data.frame(
  x = rep(1:5, 3),
  y = c(rep(0, 5), rep(1, 5), rep(2, 5)),
  height = c(0, 1, 3, 4, 0, 1, 2, 3, 5, 4, 0, 5, 4, 4, 1)
)

ggplot(d, aes(x, y, height = height, group = y)) + 
  geom_ridgeline(fill = "lightblue")

It is also possible to draw ridgelines with geom_density_ridges if we set stat = "identity". In this case, the heights are automatically scaled such that the highest ridgeline just touches the one above at scale = 1.

ggplot(d, aes(x, y, height = height, group = y)) + 
  geom_density_ridges(stat = "identity", scale = 1)

Density ridgeline plots

The geom geom_density_ridges calculates density estimates from the provided data and then plots those, using the ridgeline visualization. The height aesthetic does not need to be specified in this case.

ggplot(iris, aes(x = Sepal.Length, y = Species)) + geom_density_ridges()

There is also geom_density_ridges2, which is identical to geom_density_ridges except it uses closed polygons instead of ridgelines for drawing.

ggplot(iris, aes(x = Sepal.Length, y = Species)) + geom_density_ridges2()

The grouping aesthetic does not need to be provided if a categorical variable is mapped onto the y axis, but it does need to be provided if the variable is numerical.

# modified dataset that represents species as a number
iris_num <- transform(iris, Species_num = as.numeric(Species))

# does not work, causes error
# ggplot(iris_num, aes(x = Sepal.Length, y = Species)) + geom_density_ridges()

# works 
ggplot(iris_num, aes(x = Sepal.Length, y = Species_num, group = Species_num)) + 
  geom_density_ridges()

Trailing tails can be cut off using the rel_min_height aesthetic. This aesthetic sets a percent cutoff relative to the highest point of any of the density curves. A value of 0.01 usually works well, but you may have to modify this parameter for different datasets.

ggplot(iris, aes(x = Sepal.Length, y = Species)) + 
  geom_density_ridges(rel_min_height = 0.01)

The extent to which the different densities overlap can be controlled with the scale parameter. A setting of scale=1 means the tallest density curve just touches the baseline of the next higher one. Smaller values create a separation between the curves, and larger values create more overlap.

# scale = 0.9, not quite touching
ggplot(iris, aes(x = Sepal.Length, y = Species)) + geom_density_ridges(scale = 0.9)

# scale = 1, exactly touching
ggplot(iris, aes(x = Sepal.Length, y = Species)) + geom_density_ridges(scale = 1)

# scale = 5, substantial overlap
ggplot(iris, aes(x = Sepal.Length, y = Species)) + geom_density_ridges(scale = 5)

The scaling is calculated separately per panel, so if we facet-wrap by species each density curve exactly touches the next higher baseline. (This can be disabled by setting panel_scaling = FALSE.)

ggplot(iris, aes(x = Sepal.Length, y = Species)) + 
  geom_density_ridges(scale = 1) + facet_wrap(~Species)

Varying fill colors along the x axis

Sometimes we would like to have the area under a ridgeline not filled with a single solid color but rather with colors that vary in some form along the x axis. This effect can be achieved with the geoms geom_ridgeline_gradient and geom_density_ridges_gradient. Both geoms work just like geom_ridgeline and geom_density_ridges, except that they allow for varying fill colors. However, they do not allow for alpha transparency in the fill. For technical reasons, we can have changing fill colors or transparency but not both.

Here is a simple example of changing fill colors with geom_ridgeline_gradient:

d <- data.frame(
  x = rep(1:5, 3) + c(rep(0, 5), rep(0.3, 5), rep(0.6, 5)),
  y = c(rep(0, 5), rep(1, 5), rep(3, 5)),
  height = c(0, 1, 3, 4, 0, 1, 2, 3, 5, 4, 0, 5, 4, 4, 1))

ggplot(d, aes(x, y, height = height, group = y, fill = factor(x+y))) +
  geom_ridgeline_gradient() +
  scale_fill_viridis_d(direction = -1, guide = "none")

And here is an example using geom_density_ridges_gradient. Note that we need to map the calculated x value (stat(x)) onto the fill aesthetic, not the original temperature variable. This is the case because geom_density_ridges_gradient calls stat_density_ridges (described in the next section) which calculates new x values as part of its density calculation.

ggplot(lincoln_weather, aes(x = `Mean Temperature [F]`, y = Month, fill = stat(x))) +
  geom_density_ridges_gradient(scale = 3, rel_min_height = 0.01) +
  scale_fill_viridis_c(name = "Temp. [F]", option = "C") +
  labs(title = 'Temperatures in Lincoln NE in 2016')

Quantile lines and coloring by quantiles or probabilities

By setting the option quantile_lines = TRUE, we can make stat_density_ridges calculate the position of lines indicating quantiles. By default, three lines are drawn, corresponding to the first, second, and third quartile:

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  stat_density_ridges(quantile_lines = TRUE)

We can change the number of quantiles by specifying it via the quantiles option. Note that quantiles = 2 implies one line (the median) at the boundary between the two quantiles.

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  stat_density_ridges(quantile_lines = TRUE, quantiles = 2)

We can also specify quantiles by cut points rather than number. E.g., we can indicate the 2.5% and 97.5% tails.

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  stat_density_ridges(quantile_lines = TRUE, quantiles = c(0.025, 0.975), alpha = 0.7)

Using the geom geom_density_ridges_gradient we can also color by quantile, via the calculated stat(quantile) aesthetic. Note that this aesthetic is only calculated if calc_ecdf = TRUE.

ggplot(iris, aes(x=Sepal.Length, y=Species, fill = factor(stat(quantile)))) +
  stat_density_ridges(
    geom = "density_ridges_gradient", calc_ecdf = TRUE,
    quantiles = 4, quantile_lines = TRUE
  ) +
  scale_fill_viridis_d(name = "Quartiles")

We can use the same approach to highlight the tails of the distributions.

ggplot(iris, aes(x = Sepal.Length, y = Species, fill = factor(stat(quantile)))) +
  stat_density_ridges(
    geom = "density_ridges_gradient",
    calc_ecdf = TRUE,
    quantiles = c(0.025, 0.975)
  ) +
  scale_fill_manual(
    name = "Probability", values = c("#FF0000A0", "#A0A0A0A0", "#0000FFA0"),
    labels = c("(0, 0.025]", "(0.025, 0.975]", "(0.975, 1]")
  )

Finally, when calc_ecdf = TRUE, we also have access to a calculated aesthetic stat(ecdf), which represents the empirical cumulative density function for the distribution. This allows us to map the probabilities directly onto color.

ggplot(iris, aes(x = Sepal.Length, y = Species, fill = 0.5 - abs(0.5 - stat(ecdf)))) +
  stat_density_ridges(geom = "density_ridges_gradient", calc_ecdf = TRUE) +
  scale_fill_viridis_c(name = "Tail probability", direction = -1)

Jittering points

The stat stat_density_ridges also provides the option to visualize the original data points from which the distributions are generated. This can be done by setting jittered_points = TRUE, either in stat_density_ridges or in geom_density_ridges:

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  geom_density_ridges(jittered_points = TRUE)

Where the points are shown can be controlled with position options, e.g. “raincloud” for the raincloud effect:

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  geom_density_ridges(
    jittered_points = TRUE, position = "raincloud",
    alpha = 0.7, scale = 0.9
  )

We can also simulate a rug:

ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  geom_density_ridges(
    jittered_points = TRUE,
    position = position_points_jitter(width = 0.05, height = 0),
    point_shape = '|', point_size = 3, point_alpha = 1, alpha = 0.7,
  )

Note that we are using position_points_jitter() here, not position_jitter(). We do this because position_points_jitter() knows to jitter only the points in a ridgeline plot, without touching the density lines.

Styling the jittered points is a bit tricky but is possible with special scales provided by ggridges. First, there is scale_discrete_manual() which can be used to make arbitrary discrete scales for arbitrary aesthetics. We use it in the next example to style the point shapes. Second, there are various point aesthetic scales, such as scale_point_color_hue(). See the reference documentation for these scales for more details.

ggplot(iris, aes(x = Sepal.Length, y = Species, fill = Species)) +
  geom_density_ridges(
    aes(point_color = Species, point_fill = Species, point_shape = Species),
    alpha = .2, point_alpha = 1, jittered_points = TRUE
  ) +
  scale_point_color_hue(l = 40) +
  scale_discrete_manual(aesthetics = "point_shape", values = c(21, 22, 23))