Circular Kernel Density Plot

Plots multiples of a von Mises, wrapped Cauchy, and wrapped Normal density distribution in a circular plot

Usage

plot_density(
  x,
  bw = NULL,
  kernel = c("vonmises", "wrappedcauchy", "wrappednormal"),
  weights = NULL,
  axial = TRUE,
  n = 512L,
  norm.density = TRUE,
  kappa = NULL,
  rho = NULL,
  sd = NULL,
  c = NULL,
  fill = FALSE,
  scale = 0,
  shrink = 1,
  add = TRUE,
  main = NULL,
  labels = TRUE,
  at = seq(0, 360 - 45, 45),
  cborder = TRUE,
  grid = FALSE,
  ...
)

Arguments

x: Either an object of class "density" or a numeric vector of angles (in degrees) from which the estimate is to be computed
bw, kappa, rho, sd, c: numeric. Smoothing bandwidth expressed as the concentration parameter \(\kappa\) for the von Mises distribution, \(\rho\) for the wrapped Cauchy distribution, or \(\sigma\) for the wrapped normal distribution. Small and large values for the von Mises and wrapped normal/Cauchy distribution, respectively, gives smooth density lines. If not specified, parameter will be estimated using est.kappa() for the von Mises distribution, or set to \(p \exp(-1)\) and 1 for the wrapped Cauchy and wrapped Normal distribution (where \(p = 2\) when axial=TRUE and 1 otherwise), respectively.
kernel: character. The smoothing kernel to be used; one of "vonmises" (the default), "wrappedcauchy", "wrappednormal, for the von Mises, the wrapped Cauchy, and the wrapped Normal distribution.
weights: numeric. A vector of observation weights, of the same length as x, to give individual observations weight in the density estimate. Should sum to 1; a warning is issued if it doesn't (unless subdensity = TRUE). Defaults to equal weight 1/length(x) per observation, matching stats::density().
axial: Logical. Whether data are uniaxial (axial=FALSE) or biaxial (TRUE, the default).
n: integer. Number of equally spaced angles at which the density is to be estimated.
norm.density: logical. Normalize the density?
fill: logical. Whether to fill the density curve or draw just a line (the default)
scale: numeric. radius of plotted circle. Default is 1.1.
shrink: numeric. parameter that controls the size of the plotted function. Default is 1.
add: logical. Add to existing plot? (TRUE by default).
main: Character string specifying the title of the plot.
labels: Either a logical value indicating whether to plot labels next to the tick marks, or a vector of labels for the tick marks.
at: Optional vector of angles at which tick marks should be plotted. Set at=numeric(0) to suppress tick marks.
cborder: logical. Border of rose plot.
grid: logical. Whether a grid should be added.
...: Further graphical parameters may also be supplied as arguments.

Value

plot or calculated densities

Examples

# Filled von Mises kernel density curve inside the plot
plot_density(san_andreas$azi,
  kappa = 100,
  fill = TRUE, col = "#51127C80", border = "#51127CFF",
  grid = TRUE,
  add = FALSE
)

# Superimpose a wrapped Cauchy kernel distribution curve
plot_density(san_andreas$azi,
  rho = 0.9, kernel = "wrappedcauchy",
  fill = FALSE, col = "#FB8861FF",
  add = TRUE
)

# Superimpose a wrapped Normal kernel distribution curve
plot_density(san_andreas$azi,
  sd = 2, kernel = "wrappednormal",
  fill = FALSE, col = "#E65164FF",
  add = TRUE
)


# Superimpose a von Mises kernel density curve on a rose diagram:
rose(san_andreas$azi, grid = TRUE)
plot_density(san_andreas$azi,
  bw = 100, col = "#51127CFF",
  add = TRUE, lwd = 3
)


# Corona plot (density curve outside of a rose diagram plot):
w <- weighting(san_andreas$unc)
rose(san_andreas$azi, weights = w, dots = TRUE, stack = TRUE, dot_cex = 0.5, dot_pch = 21)
plot_density(san_andreas$azi, weights = w,
  bw = 100,
  scale = 1.1, shrink = 3, xpd = NA,
  col = "#51127CFF"
)

Usage

Arguments

Value

See also

Examples