threshold_matrix: Threshold a correlation matrix

In abnormally-distributed/rsfcNet: Analyze resting state functional connectivity networks

Description

This function is used to threshold a correlation matrix in order to prune weak connections.

Usage

threshold_matrix(c, q.thresh = 0.9, thresh)

Arguments

c	The network either as connectivity matrix or as an igraph object.
q.thresh	The quantile to threshold. Defaults to .90, such that all below the 90th quantile are set to zero.
thresh	The number (taken to be an absolute value) below which to threshold (ie, correlation or cohen's d , etc). No default. Recommend .10 for correlation or .20 for cohen's d.

Value

The pruned correlation matrix.

Author(s)

Brandon Vaughan

Examples

threshold_matrix(matrix, q=.90) # This should threshold such that only positive values are returned.
threshold_matrix(abs(matrix), q=.90) # Taking the absolute value of the matrix will allow negative correlations sufficiently strong to be included as an edge. This does not preserve signs.
threshold_matrix(abs(matrix), q=.90)*sign(matrix) This will restore the signs after thresholding if desired.

abnormally-distributed/rsfcNet documentation built on March 8, 2020, 5:32 p.m.