use_label: Assemble label and map it

In ggpmisc: Miscellaneous Extensions to 'ggplot2'

Assemble label and map it

Description

Assemble model-fit-derived text or expressions and map them to the label aesthetic.

Usage

use_label(..., labels = NULL, other.mapping = NULL, sep = "*\", \"*")

Arguments

...	character Strings giving the names of the label components in the order they will be included in the combined label.
labels	character A vector with the name of the label components. If provided, values passed through ... are ignored.
other.mapping	An unevaluated expression constructed with function aes to be included in the returned value.
sep	character A string used as separator when pasting the label components together.

Details

Statistics stat_poly_eq, stat_ma_eq, stat_quant_eq and stat_correlation return multiple text strings to be used individually or assembled into longer character strings depending on the labels actually desired. Assembling and mapping them requires verbose R code and familiarity with R expression syntax. Function use_label() automates these two tasks and accepts abbreviated familiar names for the parameters in addition to the name of the columns in the data object returned by the statistics. The default separator is that for expressions.

The statistics return variables with names ending in .label. This ending can be omitted, as well as .value for f.value.label, t.value.label, z.value.label, S.value.label and p.value.label. R2 can be used in place of rr. Furthermore, case is ignored.

Function use_label() calls aes() to create a mapping for the label aesthetic, but it can in addition combine this mapping with other mappings created with aes().

Value

A mapping to the label aesthetic and optionally additional mappings as an unevaluated R expression, built using function aes, ready to be passed as argument to the mapping parameter of the supported statistics.

Note

Function use_label() can be only used to generate an argument passed to formal parameter mapping of the statistics stat_poly_eq, stat_ma_eq, stat_quant_eq and stat_correlation.

See Also

stat_poly_eq, stat_ma_eq, stat_quant_eq and stat_correlation.

Examples

# generate artificial data
set.seed(4321)
x <- 1:100
y <- (x + x^2 + x^3) + rnorm(length(x), mean = 0, sd = mean(x^3) / 4)
my.data <- data.frame(x = x,
                      y = y * 1e-5,
                      group = c("A", "B"),
                      y2 = y * 1e-5 + c(2, 0))

# give a name to a formula
formula <- y ~ poly(x, 3, raw = TRUE)

# default label constructed by use_label()
ggplot(data = my.data,
       mapping = aes(x = x, y = y2, colour = group)) +
  geom_point() +
  stat_poly_line(formula = formula) +
  stat_poly_eq(mapping = use_label(),
               formula = formula)

# user specified label components
ggplot(data = my.data,
       mapping = aes(x = x, y = y2, colour = group)) +
  geom_point() +
  stat_poly_line(formula = formula) +
  stat_poly_eq(mapping = use_label("eq", "F"),
              formula = formula)

# user specified label components and separator
ggplot(data = my.data,
       mapping = aes(x = x, y = y2, colour = group)) +
  geom_point() +
  stat_poly_line(formula = formula) +
  stat_poly_eq(mapping = use_label("R2", "F", sep = "*\" with \"*"),
               formula = formula)

# combine the mapping to the label aesthetic with other mappings
ggplot(data = my.data,
       mapping = aes(x = x, y = y2)) +
  geom_point(mapping = aes(colour = group)) +
  stat_poly_line(mapping = aes(colour = group), formula = formula) +
  stat_poly_eq(mapping = use_label("grp", "eq", "F",
                                   aes(grp.label = group)),
              formula = formula)

# combine other mappings with default labels
ggplot(data = my.data,
       mapping = aes(x = x, y = y2)) +
  geom_point(mapping = aes(colour = group)) +
  stat_poly_line(mapping = aes(colour = group), formula = formula) +
  stat_poly_eq(mapping = use_label(aes(colour = group)),
              formula = formula)

# example with other available components
ggplot(data = my.data,
       mapping = aes(x = x, y = y2, colour = group)) +
  geom_point() +
  stat_poly_line(formula = formula) +
  stat_poly_eq(mapping = use_label("eq", "adj.R2", "n"),
               formula = formula)

# multiple labels
ggplot(data = my.data,
       mapping = aes(x, y2, colour = group)) +
  geom_point() +
  stat_poly_line(formula = formula) +
  stat_poly_eq(mapping = use_label("R2", "F", "P", "AIC", "BIC"),
               formula = formula) +
  stat_poly_eq(mapping = use_label(c("eq", "n")),
               formula = formula,
               label.y = "bottom",
               label.x = "right")

# quantile regression
ggplot(data = my.data,
       mapping = aes(x, y)) +
  stat_quant_band(formula = formula) +
  stat_quant_eq(mapping = use_label("eq", "n"),
                formula = formula) +
  geom_point()

# major axis regresion
ggplot(data = my.data, aes(x = x, y = y)) +
  stat_ma_line() +
  stat_ma_eq(mapping = use_label("eq", "n")) +
  geom_point()

# correlation
ggplot(data = my.data,
       mapping = aes(x = x, y = y)) +
  stat_correlation(mapping = use_label("r", "t", "p")) +
  geom_point()

ggpmisc documentation built on June 28, 2024, 1:07 a.m.