[R] pca

[Chuck Cleland]

Prof Brian Ripley wrote:
> On Sun, 14 Dec 2003, Douglas Trainor wrote:
> 
> 
>>Somewhere along the line, you have been confused.
>>You're in good company though.  Factor analysis 
>>and PCA are different entities entirely.
> 
> 
> Not in SPSS, where the same command is used for both (although we were not 
> told anything like enough about what was done).
> 
> *However* I don't know what is meant by `pca in R'. Standard R contains
> princomp() and prcomp() to do PCA: it appears that pca() is in the
> orphaned package multiv. Why not use the standard functions?  Also, those
> non-default arguments would appear to be very unusual indeed for PCA, and
> do not correspond to a *correlation* matrix.

   It looks like the original poster might have used the pca() in 
the pcurve package, which seems to require a data matrix, not a 
correlation matrix.

>>Any clues about this diference?
> 
> 
> User error looks likely.  Please seek out local statistical expertise.
> 
> 
>>>Dear listmates, i've done a pca analisys in R (1.8 v.) 
>>>with the command
>>>
>>>pca(Matrix, cent=FALSE, scle=FALSE)  
>>>
>>>I have obtained a v matrix very different from the  
>>>component matrix resulted by a factor analysis in SPSS, 
>>>unrotated and with a extraction from a correlation 
>>>matrix. Any clues about this diference?

-- 
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894