Sparklines

Draw a time series small enough to place in a sentence, table cell, or margin. Tufte describes these word-sized graphics in Beautiful Evidence.

Usage

sparkline(
  values,
  index = seq_along(values),
  band = c(0.25, 0.75),
  band_fill = "grey90",
  extremes = TRUE,
  last_point = TRUE,
  label = TRUE,
  accuracy = 0.1,
  big.mark = ",",
  colour = "grey15",
  linewidth = 0.3,
  extreme_colours = c("#4a6b82", "#a1483c")
)

Arguments

values

Numeric vector of values, in order.

index

Optional numeric vector of positions. Defaults to seq_along(values).

band

Either NULL for no band, or a length-2 numeric vector of quantiles defining the normal range. Defaults to c(0.25, 0.75).

band_fill

Fill colour of the normal-range band.

extremes

Logical. Mark the minimum and maximum with dots?

last_point

Logical. Mark the final value with a dot?

label

Logical. Print the final value at the right-hand end?

accuracy

Rounding for the printed value, passed to label_number(). Defaults to 0.1; use 1 for counts.

big.mark

Thousands separator for the printed value. Defaults to a comma, since the scales default is a space and reads oddly at sparkline size.

colour

Line colour. Defaults to a near-black grey.

linewidth

Line width. Defaults to 0.3, a hairline.

extreme_colours

Length-2 vector of colours for the minimum and maximum dots.

Value

A ggplot object with no axes, sized to be printed small.

Details

The default grey band marks the interquartile range. Dots identify the minimum and maximum, and a label gives the final value. There are no axes, so make sure the surrounding text supplies the context readers need.

See also

sparklines() for many series at once, sparkline_grob() to embed one in other graphics.

Examples

set.seed(1)
sparkline(cumsum(rnorm(80)))