Compute convex scagnostic measures

Compute convex scagnostic measures

sc_skinny(x, y)

# S3 method for scree
sc_skinny(x, y = NULL)

# S3 method for default
sc_skinny(x, y)

# S3 method for list
sc_skinny(ahull)

Examples

  require(ggplot2)
  require(tidyr)
  require(dplyr)
  data(anscombe_tidy)
  anscombe_tidy <- anscombe %>%
  pivot_longer(cols = everything(),
    names_to = c(".value", "set"),
    names_pattern = "(.)(.)")
  ggplot(anscombe_tidy, aes(x=x, y=y)) +
    geom_point() +
    facet_wrap(~set, ncol=2, scales = "free")

  a_s <- scree(anscombe$x1, anscombe$y1)
  sc_skinny(a_s)
#> [1] 0.3616838
  a_s <- scree(anscombe$x2, anscombe$y2)
  sc_skinny(a_s)
#> [1] 0.4416274
  a_s <- scree(anscombe$x3, anscombe$y3)
  sc_skinny(a_s)
#> [1] 0.5229884
  a_s <- scree(anscombe$x4, anscombe$y4)
#> Warning: success: collinearities reduced through jitter
#> Warning: dataset started with 3 colinear points, jitter added!
  sc_skinny(a_s)
#> [1] 0.4616928