Displaying 1 result from an estimated 1 matches for "pcor2".
Did you mean:
pcor
2000 Feb 25
0
Summary: Partial correlation coefficients in R. Thanks everybody!
...y John Logsdon, among others:
The partial correlation coefficients are the negative
scaled off-diagonal inverse variance. So compute the variance-covariance
matrix, invert, scale to the diagonal and negate and you have it.
And an implementation by Martyn Plummer (here, too, I received several):
pcor2 <- function(x){
conc <- solve(var(x))
resid.sd <- 1/sqrt(diag(conc))
pcc <- - sweep(sweep(conc, 1, resid.sd, "*"), 2, resid.sd, "*")
return(pcc)
}
This is the version I'm using now, together with a test for significance of
each co...