scagnostics: Calculcate scagnostics for pairs of variables

View source: R/scagnostics.R

scagnosticsR Documentation

Calculcate scagnostics for pairs of variables

Description

Scagnostics (scatterplot diagnostics) summarize potentially interesting patterns in 2d scatterplots.

Usage

scagnostics(x, ...)
scagnosticsOutliers(scagnostics)
scagnosticsExemplars(scagnostics)
scagnosticsGrid(scagnostics)

Arguments

x

object to calculate scagnostics on: a vector, a matrix or a data.frame

...

...

scagnostics

objects returned from the scagnostics function

Value

scagnostics returns a vector (for a pair) or a matrix consisting of scagnostics (rows) by variable pairs (columns).

scagnostics.outliers and scagnostics.exemplars return a logical vector.

scagnostics.grid returns a data frame with columns x and y which as a pair define the index of variables corresponding to the entries in the scagnostics matrix. Hence it has as many rows as there are columns in the scagnostics matrix and each row defines one pair of variables.

Author(s)

Lee Wilkinson <leland.wilkinson@gmail.com>, Anushka Anand (University of Illinois at Chicago), R package: Simon Urbanek <simon.urbanek@r-project.org>

References

Wilkinson L., Anand, A., and Grossman, R. (2006). High-Dimensional visual analytics: Interactive exploration guided by pairwise views of point distributions. IEEE Transactions on Visualization and Computer Graphics, November/December 2006 (Vol. 12, No. 6) pp. 1363-1372.

Wilkinson L., Anand, A., and Grossman, R. (2005). Graph-Theoretic Scagnostics. Proceedings of the 2005 IEEE Symposium on Information Visualization, p. 21.

https://www.cs.uic.edu/~wilkinson/

Examples

# tesing various forms
scagnostics(1:10, 1:10)
scagnostics(rnorm(100), rnorm(100))
scagnostics(as.matrix(mtcars))

# more real use on a dataset
s <- scagnostics(mtcars)

# look at outliers
o <- scagnosticsOutliers(s)
o[o]
# one outlier, let's plot it
g <- scagnosticsGrid(s)
go <- g[o,]
plot(mtcars[[go$x]], mtcars[[go$y]], pch=19,
     xlab=names(mtcars)[go$x], ylab=names(mtcars)[go$y])

# find and plot exemplars
e <- scagnosticsExemplars(s)
e[e]
ge <- g[e,]
par(mfrow = c(2,2))
for (i in 1:dim(ge)[1])
  plot(mtcars[[ge$x[i]]], mtcars[[ge$y[i]]], pch=19,
       xlab=names(mtcars)[ge$x[i]], ylab=names(mtcars)[ge$y[i]])

scagnostics documentation built on March 18, 2022, 7:17 p.m.