simplexplot {archetypes}

Simplex visualization

Description

The stochastic nature of the alpha coefficients implies that they exist on a standard (K-1)-simplex with the K archetypes Z as the corners, and the coefficients as the coordinate with respect to these corners. A standard simplex can be projected to two dimensions via a skew orthogonal projection, where all the vertices of the simplex are shown on a circle connected by edges. The individual alpha coefficients can be then projected into this circle.

Usage

simplexplot(object, radius = 10, order = NULL, labels_cex = 1,
  labels = NULL, show_labels = TRUE, points_col = "#00000044",
  points_pch = 19, points_cex = 1, projection = simplex_projection,
  show_points = TRUE, show_circle = TRUE, circle_col = "lightgray",
  show_edges = TRUE, edges_col = "lightgray", show_direction = FALSE,
  direction_length = 1, directions_col = points_col, ...)

Arguments

object

An archetypes object

radius

Radius of the projection

order

Order of the archetypes

labels_cex

Label expansion

labels

Labels

show_labels

Show labels

points_col

Color of the points

points_pch

Plot character of the points

points_cex

Character expansion of the points

projection

Projection function; see archmap_projections

show_points

Show the points

show_circle

Show the circle

circle_col

Color of the circle

show_edges

Show the edges

edges_col

Color of the edges

direction_length

Expansion of the direction pointers

directions_col

Color of the direction pointers

show_direction

Show direction pointers

...

Additional arguments; currently ignored

Value

Invisible list of all computed components needed for the simplex visualization.

References

See Section 6 in "Probabilistic Archetypal Analysis" by Seth and Eugster (2014), http://arxiv.org/abs/1312.7604.

Examples

### This example reproduces parts of the Figure 7 shown in
  ### "Probabilistic Archetypal Analysis" by Seth and Eugster (2014)

  data("toy", package = "archetypes")

  suppressWarnings(RNGversion("3.5.0"))
  set.seed(1234); a3 <- archetypes(toy, k = 3)
  set.seed(1237); a4 <- archetypes(toy, k = 4)
  set.seed(1238); a5 <- archetypes(toy, k = 5)

  simplexplot(a3)
  simplexplot(a3, show_direction = TRUE, show_points = FALSE)
  simplexplot(a4, projection = tspsimplex_projection)
  simplexplot(a5, show_direction = TRUE, show_points = FALSE,
    direction_length = 2, directions_col = "black")