scaggram: Visualization of scagnostics results in matrix graphics
In mbgraphic: Measure Based Graphic Selection

Description Usage Arguments Details Note Author(s) See Also Examples

Scaggram are thought as a generalization of corrgrams. It is possible to use up to three arbitrary scagnostics at the same time. Each of the measures is represented by one of the colors red, green, or blue. The resulting color of a cell (representing a pair of variables) in the scaggram is given by the mixture of the (up to) three colors according to the values of the scagnostics.

1	scaggram(sdfdata,select=1,add=FALSE,label=FALSE,order= FALSE)

`sdfdata`	A list of class `"sdfdata"`.
`select`	Numeric or character vector of length `1, 2` or `3`. Give the column numbers of the measures or their names. Default is `1`.
`add`	Logical. Which additional features should be added to the plot? `FALSE` for none, `"splom"` for scatterplots and `"glyphs"` for star glyphs of all scagnostics stored in `sdfdata$sdf`.
`label`	Should the names of the variables be plotted? Default is `FALSE`.
`order`	Should the variables be reordered? Default is `FALSE`. Use `2` or `"quicksort"` for quick reorering with function `sdf_quicksort`, `3` or `"All scagnostics"` for reordering with function `sdf_sort` based on all scagnosticsand `4` or `"Selected scagnostics"` for reordering with `sdf_sort` based on the selected scagnostics.

Given three scagnostics with values s1, s2, s3 describing a pair of variables, then the color of the respective cell of the scaggram is determined by rgb(s1,s2,s3, alpha = max(s1,s2,s3)).

If glyphs or a splom are added by setting add, the part above the diagonal is used for the additional feature and the color of the cells is drawn with a transparency of max(s1,s2,s3,0.2)).

The scagnostics need to range between 0 and 1.

For a good and informative color result, it is required that the scagnostics used have similar scales. Otherewise the result can be strongly influenced by single measures.

Katrin Grimm

iascaggram, sdf, scag2sdf, sdf_sort, sdf_quicksort

rgb

data(Election2005)
# Results of the election
## Not run: 
sdfres <- sdf(Election2005[,41:70])
# Use scagnostics "Outlying", "Clumpy" and "Monotonic"
scaggram(sdfres,select=c(1,3,9))

## End(Not run)