Package

ggExtra

Author

Dean Attali

Adding marginal histograms with ggExtra

The ggMarginal function of the ggExtra package allows adding marginal histograms to an existing scatter plot. For that purpose you will need to store the scatter plot made with ggplot2 inside a variable and pass it to ggMarginal, specifying type = "histogram".

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Plot the scatter plot with marginal histograms
ggMarginal(p, type = "histogram")

Scatter plot with marginal histograms in ggplot2

Histogram only on the horizontal axis

The margins argument can be used to set only one of the marginal histograms, as the X-axis histogram.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Horizontal marginal histogram
ggMarginal(p, type = "histogram", 
           margins = "x")

Marginal histogram on the X-axis

Histogram only on the vertical axis

If you prefer showing only the Y-axis histogram set margins = "y".

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Vertical marginal histogram
ggMarginal(p, type = "histogram", 
           margins = "y")

Scatter plot in ggplot with marginal histogram on the vertical axis

Relative size

The size argument modifies the relative size between the histogram and the scatter plot. The default value is 5 (scatter plot 5 times bigger than the histograms).

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Changing the relative size
ggMarginal(p, type = "histogram", 
           size = 3)

Marginal histograms relative size

Arguments to be passed to geom_boxplot

You can pass the arguments of geom_boxplot to the function. If you want to set arguments for each histogram pass lists to xparams and yparams.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Bin width customization
ggMarginal(p, type = "histogram", 
           binwidth = 4)

Customization of the ggplot2 marginal histograms

Histogram with density lines

If you want to overlay density lines to the histograms set type = "densigram".

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Densigram
ggMarginal(p, type = "densigram")

Marginal histograms with density, also called densigram in ggplot2

Color customization

ggplot2 marginal histograms color

Fill color

Change the fill color of the histograms with fill.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Changing the fill color
ggMarginal(p, type = "histogram", 
           fill = 4)

ggplot marginal histogram border color

Border color

You can also customize the border color with col argument.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Fill and border color of the marginal histograms
ggMarginal(p, type = "histogram", 
           fill = "white",
           col = 4)

Scatter plot with marginal histograms in R

Arguments for each histogram

If you want more customization, pass a list or arguments for each axis with xparams and yparams.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist)) +
  geom_point()

# Arguments for each marginal histogram
ggMarginal(p, type = "histogram", 
           xparams = list(fill = 4),
           yparams = list(fill = 3))

Marginal histograms by group in R ggplot2

Histograms by group

Finally, if your data set contains a grouping variable you can fill the marginal histograms by group.

# install.packages("ggplot2")
# install.packages("ggExtra")
library(ggplot2)
library(ggExtra)

# Sample grouping variable
cars$group <- c(rep("A", 25), rep("B", 25))

# Save the scatter plot in a variable
p <- ggplot(cars, aes(x = speed, y = dist, color = group)) +
  geom_point()

# Marginal histograms by group
ggMarginal(p, type = "histogram", 
           groupColour = TRUE,
           groupFill = TRUE)

See also