Hi,
The help file on princomp says that 'The signs of the columns of the
loadings and scores are arbitrary, and so may differ between differnt
programs for PCA, and even between different builds of R'. Does anyone
know
1. Whether this applys to all functions like svd, eigen ... etc?
2. During a leave-one-out cross step, I notice the loadings also change
sige when a different observation is held out. This causes the sign change
of principal component score of the held-out observation. Since I use the
same version of R, same procedure, and same program, to make the
leave-one-out PC scores consistent, does that mean I have the take the
absolute value of these scores before analyzing them?
Thanks in advance for any help!
Anne
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
