hfgolino/EGAnet

Adaptive alpha p-value in UVA

Closed this issue · 2 comments

I'm getting the following error message when I try to run UVA:

Error in redundancy.process(data = data, cormat = cormat, n = n, model = model, :
"alpha" and "adapt" have been removed from UVA. Please use "threshold"

Is it no longer possible to use "adapt" at all?

At least in the updated package, this functionality has been removed in favor of "threshold" which is demonstrated to be more accurate than "adapt". "adapt" tended to overestimate the number of locally dependent pairs.

If you're looking to be more lenient, you can try a threshold = 0.15 or 0.20

Linking archived {EGAnet} version 1.2.0 which has this capability still: https://cran.r-project.org/src/contrib/Archive/EGAnet/EGAnet_1.0.0.tar.gz