method_kde {ggdensity}

Bivariate kernel density HDR estimator

Description

Function used to specify bivariate kernel density estimator for get_hdr() and layer functions (e.g. geom_hdr()).

Usage

method_kde(h = NULL, adjust = c(1, 1))

Arguments

h

Bandwidth (vector of length two). If NULL, estimated using MASS::bandwidth.nrd().

adjust

A multiplicative bandwidth adjustment to be used if 'h' is 'NULL'. This makes it possible to adjust the bandwidth while still using the a bandwidth estimator. For example, adjust = 1/2 means use half of the default bandwidth.

Details

For more details on the use and implementation of the ⁠method_*()⁠ functions, see vignette("method", "ggdensity").

Examples

set.seed(1)
df <- data.frame(x = rnorm(1e3, sd = 3), y = rnorm(1e3, sd = 3))

ggplot(df, aes(x, y)) +
  geom_hdr(method = method_kde()) +
  geom_point(size = 1)

# The defaults of `method_kde()` are the same as the estimator for `ggplot2::geom_density_2d()`
ggplot(df, aes(x, y)) +
  geom_density_2d_filled() +
  geom_hdr_lines(method = method_kde(), probs = seq(.1, .9, by = .1)) +
  theme(legend.position = "none")

# The bandwidth of the estimator can be set directly with `h` or scaled with `adjust`
ggplot(df, aes(x, y)) +
  geom_hdr(method = method_kde(h = 1)) +
  geom_point(size = 1)

ggplot(df, aes(x, y)) +
  geom_hdr(method = method_kde(adjust = 1/2)) +
  geom_point(size = 1)

# Can also be used with `get_hdr()` for numerical summary of HDRs
res <- get_hdr(df, method = method_kde())
str(res)