Good day all.
This is to thank all those who have helped in fixing this problem. Starting
with a text book was indeed a problem, however, that gave me a clue of what
I was looking for. This, with your contributions added to other materials I
got on the net, put me on the right track. Thank you so much.
Warmest regards
Ogbos
On 31 January 2010 14:07, S Ellison <S.Ellison@lgc.co.uk> wrote:
> I doubt you will get a useful answer as to why you cannot reporduce an
> example from a book - especially one you didn't cite!
>
> R has prcomp and princomp in the base package, and as others have
> pointed out, pca is implementd in other packages. We used ade4 a lot; it
> has quite nice default graphics.
>
> If you want to calculate principal components manually, you probably
> need to look at eigen() and remember a) in PCA, the vectors are usually
> given in order of decreasing eigenvalue and b) eigenvectors are not
> generally unique, especially as to sign. Different PCA applciations may
> give you eigenvalues of differing sign.
>
> Also, eigenvector solutions are not the only way of obtaining
> eigenvalues; efficient solutions for a limited number of PC's also
> exist, particularly the NIPALS algorithm
>
>
> read your text book to the point where it mentions the eigenvectors of
> the ccorrelation (or covariance) matrix
> >>> ogbos okike <ogbos.okike@gmail.com> 01/30/10 7:09 PM
>>>
> Hi,
> I am learning how to do principal component analysis in R. However,
> since I
> am family with only a few built-in functions like prcomp, sd, cor, I
> started
> manually with examples in text books while trying to use the few
> functions I
> know to manipulate what they have in the text. From the example in the
> text
> I obtained a data set. Using cor and cov, I calculated the correlation
> and
> covariance of the data frame. I equally calculated standardized data as
> they
> did. They plotted a graph of the standardized X against Y, X against Z
> and Z
> against Y. I tried to plot the same graph in R but could not fit the
> First
> Principal Component as they did.
> I will be glad if anybody would be good enough as to guide me on how to
> fit
> this first (and probably second, third) principal component (s). As a
> begginer, I would appreciate any additional information on how to
> proceed
> with pca in R.
> Thank you.
> Ogbos
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>
> *******************************************************************
> This email and any attachments are confidential. Any u...{{dropped:12}}
