hmm... isn't it just
recoverfrompca <- function(pcaobject, pcascores) {
pcascores %*% t(pcaobject$loadings)
}
???
Alexander Sirotkin wrote:
> Hi.
>
> I'm trying to find a simple way to ccalculate original data from
> PCA components using S-Plus (or R).
>
> I have smething like :
>
> p <- princomp(x)
>
> pca <- predict.princomp(p, x_new)
>
> And now... I believe there should be a simple way to comvert
> "pca" back into "x_new", but ...
>
> Any help would be appreciated...
>
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
