fdg: Force directed graph visualization of associations between...
In matie: Measuring Association and Testing Independence Efficiently

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

Constructs and draws a force directed graph using an "all pairs" association matrix to drive attraction in the graph. This adjacency matrix can represent linear or non-linear associations.

1	fdg(dataSet, dataName, method, cutoff, dim)

`dataSet`	a data frame
`dataName`	a string naming the dataSet. (set to NULL if unknown)
`method`	one of "A" for non-linear associations or "Rsq" for linear associations. Defaults to "A".
`cutoff`	a real number between 0 and 1, used to remove edges from the graph with association < cutoff. Defaults to `0.1`
`dim`	graph layouts can be generated in 2 or 3 dimensions. note that `rgl` must be installed for viewing 3 dimensional layouts.

The igraph package is used to generate, layout and display the force directed graph. The layout employed is layout.kamada.kawai.

Returns nothing.

All non numeric columns and non complete rows are stripped from the data set before the adjacency matrix is computed.

Ben Murrell, Dan Murrell & Hugh Murrell.

Discovering general multidimensional associations, http://arxiv.org/abs/1303.1828

ma cor agram

    data(baseballData)
    fdg(baseballData,dataName="baseballData",method="A",cutoff=0.35,dim=2)
    
    # if you have rgl available then try dim=3 
    # fdg(baseballData,dataName="baseballData",method="Rsq",cutoff=0.15,dim=3)

[1] "Processing 11 variables. 55 pairs"
[1] "1 variable(s) complete"
[1] "2 variable(s) complete"
[1] "3 variable(s) complete"
[1] "4 variable(s) complete"
[1] "5 variable(s) complete"
[1] "6 variable(s) complete"
[1] "7 variable(s) complete"
[1] "8 variable(s) complete"
[1] "9 variable(s) complete"
[1] "10 variable(s) complete"
Warning message:
In layout_with_kk(list(11, FALSE, c(1, 3, 4, 5, 8, 3, 4, 5, 3, 4,  :
  Argument `coolexp' is deprecated and has no effect