R Graphics · Density Plot

Exploring Distributions with Density Plots in R

Move from a basic kernel-density estimate to grouped, filled, bandwidth-adjusted, and ggplot2 density graphics.

6 examplesReproducible R code

Before you begin

Choose the variation that serves the analysis

Density plots estimate the shape of a continuous distribution. Their smooth curves make peaks, spread, skewness, and group differences easier to see than a table of values.

Examples: 06
Levels: Beginner + Intermediate

Example 01Base RBeginner

Draw a Basic Density Curve

A density curve is a smoothed estimate of where observations are concentrated.

density() calculates the estimate and plot() displays it. The total area under the curve equals one.

RRun this code

set.seed(42)
x <- rnorm(200)
plot(density(x), lwd = 3, col = "#2563eb",
     main = "Basic density plot", xlab = "Value")

Example 02Base RBeginner

Fill the Area Below the Curve

A translucent fill emphasizes the overall shape without hiding the curve boundary.

Draw the estimate first, use polygon() for the fill, and redraw the line on top for a crisp edge.

RRun this code

d <- density(rnorm(200))
plot(d, type = "n", main = "Filled density")
polygon(d, col = adjustcolor("#2563eb", .25), border = NA)
lines(d, col = "#2563eb", lwd = 3)

Example 03Base RBeginner

Compare Smoothing Bandwidths

Bandwidth controls how strongly the density estimate is smoothed.

Small bandwidths reveal local detail but may be noisy; large bandwidths produce a simpler, smoother shape.

RRun this code

x <- faithful$eruptions
plot(density(x, adjust = .5), col = "#2563eb", lwd = 2)
lines(density(x, adjust = 1.5), col = "#dc2626", lwd = 2)
legend("topright", c("Less smooth", "More smooth"),
       col = c("#2563eb", "#dc2626"), lwd = 2, bty = "n")

Example 04Base RIntermediate

Compare Several Groups

Overlaid density curves compare distributional shapes across categories.

Use a shared axis and distinct colors. Transparency or direct labels help avoid ambiguity where curves overlap.

RRun this code

cols <- c("#2563eb", "#0891b2", "#f59e0b")
sp <- levels(iris$Species)
plot(density(iris$Sepal.Length[iris$Species == sp[1]]),
     xlim = range(iris$Sepal.Length), col = cols[1], lwd = 2)
for (i in 2:3) lines(density(iris$Sepal.Length[iris$Species == sp[i]]),
                      col = cols[i], lwd = 2)
legend("topright", sp, col = cols, lwd = 2, bty = "n")

Example 05Base RIntermediate

Add the Observed Values

A rug plot displays individual observations along the axis beneath the density curve.

rug() reconnects the smooth estimate to the sample and can reveal gaps or clusters hidden by smoothing.

RRun this code

x <- faithful$waiting
plot(density(x), col = "#2563eb", lwd = 3,
     main = "Density with observed values")
rug(x, col = adjustcolor("#172554", .45))

Example 06ggplot2Intermediate

Create a Density Plot with ggplot2

ggplot2 maps grouping and fill aesthetics to layered density curves.

geom_density() supports transparent group fills and integrates with a consistent grammar for labels and themes.

RRun this code

library(ggplot2)
ggplot(iris, aes(Sepal.Length, fill = Species)) +
  geom_density(alpha = .3) +
  labs(title = "Sepal-length distributions", x = "Centimetres") +
  theme_minimal()