Adaptive alpha p-value in UVA
Closed this issue · 2 comments
kjwalter89 commented
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?
AlexChristensen commented
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
AlexChristensen commented
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