Skip to contents

Formula interface to geom_smooth()

Source: R/gf_functions.R
gf_smooth.Rd

LOESS and linear model smoothers in ggformula.

Usage

gf_smooth(
  object = NULL,
  gformula = NULL,
  data = NULL,
  ...,
  method = "auto",
  formula = y ~ x,
  se = FALSE,
  method.args,
  n = 80,
  span = 0.75,
  fullrange = FALSE,
  level = 0.95,
  xlab,
  ylab,
  title,
  subtitle,
  caption,
  geom = "smooth",
  stat = "smooth",
  position = "identity",
  show.legend = NA,
  show.help = NULL,
  inherit = TRUE,
  environment = parent.frame()
)

gf_lm(
  object = NULL,
  gformula = NULL,
  data = NULL,
  ...,
  alpha = 0.3,
  linewidth = 1,
  linetype,
  lm.args = list(),
  interval = "none",
  level = 0.95,
  fullrange = TRUE,
  xlab,
  ylab,
  title,
  subtitle,
  caption,
  geom = "lm",
  stat = "lm",
  position = "identity",
  show.legend = NA,
  show.help = NULL,
  inherit = TRUE,
  environment = parent.frame()
)

Arguments

object

When chaining, this holds an object produced in the earlier portions of the chain. Most users can safely ignore this argument. See details and examples.

gformula

A formula with shape y ~ x. Faceting can be achieved by including | in the formula.

data

A data frame with the variables to be plotted.

...

Additional arguments. Typically these are (a) ggplot2 aesthetics to be set with attribute = value, (b) ggplot2 aesthetics to be mapped with attribute = ~ expression, or (c) attributes of the layer as a whole, which are set with attribute = value.

method

Smoothing method (function) to use, accepts either NULL or a character vector, e.g. "lm", "glm", "gam", "loess" or a function, e.g. MASS::rlm or mgcv::gam, stats::lm, or stats::loess. "auto" is also accepted for backwards compatibility. It is equivalent to NULL.

For method = NULL the smoothing method is chosen based on the size of the largest group (across all panels). stats::loess() is used for less than 1,000 observations; otherwise mgcv::gam() is used with formula = y ~ s(x, bs = "cs") with method = "REML". Somewhat anecdotally, loess gives a better appearance, but is \(O(N^{2})\) in memory, so does not work for larger datasets.

If you have fewer than 1,000 observations but want to use the same gam() model that method = NULL would use, then set method = "gam", formula = y ~ s(x, bs = "cs").

formula

Formula to use in smoothing function, eg. y ~ x, y ~ poly(x, 2), y ~ log(x). NULL by default, in which case method = NULL implies formula = y ~ x when there are fewer than 1,000 observations and formula = y ~ s(x, bs = "cs") otherwise.

se

Display confidence band around smooth? (TRUE by default, see level to control.)

method.args

List of additional arguments passed on to the modelling function defined by method.

n

Number of points at which to evaluate smoother.

span

Controls the amount of smoothing for the default loess smoother. Smaller numbers produce wigglier lines, larger numbers produce smoother lines. Only used with loess, i.e. when method = "loess", or when method = NULL (the default) and there are fewer than 1,000 observations.

fullrange

If TRUE, the smoothing line gets expanded to the range of the plot, potentially beyond the data. This does not extend the line into any additional padding created by expansion.

level

Level of confidence band to use (0.95 by default).

xlab

Label for x-axis. See also gf_labs().

ylab

Label for y-axis. See also gf_labs().

title, subtitle, caption

Title, sub-title, and caption for the plot. See also gf_labs().

geom

A character string naming the geom used to make the layer.

stat

A character string naming the stat used to make the layer.

position

Either a character string naming the position function used for the layer or a position object returned from a call to a position function.

show.legend

A logical indicating whether this layer should be included in the legends. NA, the default, includes layer in the legends if any of the attributes of the layer are mapped.

show.help

If TRUE, display some minimal help.

inherit

A logical indicating whether default attributes are inherited.

environment

An environment in which to look for variables not found in data.

alpha

Opacity (0 = invisible, 1 = opaque).

linewidth

A numerical line width or a formula used for mapping linewidth.

linetype

A linetype (numeric or "dashed", "dotted", etc.) or a formula used for mapping linetype.

lm.args

A list of arguments to stats::lm().

interval

One of "none", "confidence" or "prediction".

Value

a gg object

Specifying plot attributes

Positional attributes (a.k.a, aesthetics) are specified using the formula in gformula. Setting and mapping of additional attributes can be done through the use of additional arguments. Attributes can be set can be set using arguments of the form attribute = value or mapped using arguments of the form attribute = ~ expression.

In formulas of the form A | B, B will be used to form facets using ggplot2::facet_wrap() or ggplot2::facet_grid(). This provides an alternative to gf_facet_wrap() and gf_facet_grid() that is terser and may feel more familiar to users of lattice.

Evaluation

Evaluation of the ggplot2 code occurs in the environment of gformula. This will typically do the right thing when formulas are created on the fly, but might not be the right thing if formulas created in one environment are used to create plots in another.

See also

ggplot2::geom_smooth(), gf_spline()

Examples

gf_smooth()
#> gf_smooth() uses 
#>     * a formula with shape y ~ x. 
#>     * geom:  smooth 
#>     * stat:  smooth 
#>     * key attributes:  method = "auto", formula = y ~ x, se = FALSE, method.args,
#>                    n = 80, span = 0.75, fullrange = FALSE,
#>                    level = 0.95
#> 
#> For more information, try ?gf_smooth
gf_lm()
#> gf_lm() uses 
#>     * a formula with shape y ~ x. 
#>     * geom:  lm 
#>     * stat:  lm 
#>     * key attributes:  alpha = 0.3, linewidth = 1, linetype, lm.args = list(),
#>                    interval = "none", level = 0.95,
#>                    fullrange = TRUE
#> 
#> For more information, try ?gf_lm
gf_smooth(births ~ date, color = ~wday, data = mosaicData::Births78)
#> `geom_smooth()` using method = 'loess'

gf_smooth(births ~ date,
  color = ~wday, data = mosaicData::Births78,
  fullrange = TRUE
)
#> `geom_smooth()` using method = 'loess'
#> Warning: Removed 16 rows containing missing values or values outside the scale range
#> (`geom_smooth()`).

gf_smooth(births ~ date,
  color = ~wday, data = mosaicData::Births78,
  show.legend = FALSE, se = FALSE
)
#> `geom_smooth()` using method = 'loess'

gf_smooth(births ~ date,
  color = ~wday, data = mosaicData::Births78,
  show.legend = FALSE, se = TRUE
)
#> `geom_smooth()` using method = 'loess'

gf_lm(length ~ width,
  data = mosaicData::KidsFeet,
  color = ~biggerfoot, alpha = 0.2
) |>
  gf_point()

gf_lm(length ~ width,
  data = mosaicData::KidsFeet,
  color = ~biggerfoot, fullrange = FALSE, alpha = 0.2
)

gf_point()
#> gf_point() uses 
#>     * a formula with shape y ~ x. 
#>     * geom:  point 
#>     * key attributes:  alpha, color, size, shape, fill, group, stroke
#> 
#> For more information, try ?gf_point
gf_lm(length ~ width,
  color = ~sex, data = mosaicData::KidsFeet,
  formula = y ~ poly(x, 2), linetype = "dashed"
) |>
  gf_point()

gf_lm(length ~ width,
  color = ~sex, data = mosaicData::KidsFeet,
  formula = log(y) ~ x, backtrans = exp
) |>
  gf_point()


gf_lm(hwy ~ displ,
  data = mpg,
  formula = log(y) ~ poly(x, 3), backtrans = exp,
  interval = "prediction", fill = "skyblue"
) |>
  gf_lm(
    formula = log(y) ~ poly(x, 3), backtrans = exp,
    interval = "confidence", color = "red"
  ) |>
  gf_point()


  clotting <- data.frame(
   u = c(5,10,15,20,30,40,60,80,100),
   lot1 = c(118,58,42,35,27,25,21,19,18),
   lot2 = c(69,35,26,21,18,16,13,12,12))
  gf_point(lot1 ~ u, data = clotting) |>
    gf_smooth(formula = y ~ log(x), method = "glm",
              method.args = list(family = Gamma))

  gf_point(lot2 ~ u, data = clotting) |>
    gf_smooth(formula = y ~ log(x), color = "red", method = "glm",
              method.args = list(family = Gamma))