R help - Jul 2010 - Histogram Principal component analysis in R

Hi,

I am trying to do a Principal component analysis on histogram data.
Basically, I have a group of subjects and for each of them, I have a column
of bin-counts (vis-a-vis intervals) and a corresponding column of
frequencies (or normalized frequencies). The bin counts are the same for all
the subjects. I also have a group-averaged histogram (with the same bin
counts and a column of frequencies) reflecting the overall group behavior.

I would like to apply PCA on these histograms (specifically the group
averaged one) to come up with the first few eigenvectors/PC scores that
would summarize the information contained in the histograms. Some medical
papers have done this (Dehmeshki J, Ruto AC, Arridge S...(2001) : Analysis
Of MTR histograms in multiple sclerosis using PC and muliple LDA") maybe
using some medical softwares. However, It would be great to know if there
are any R functions that can do this.
Thanks and regards,
Dhiman Bhadra



Dhiman Bhadra
Graduate Assistant.
Department of Statistics.
University of Florida.
Gainesville, Florida : 32611

	[[alternative HTML version deleted]]

Hello Dhiman ,

I have never tried doing it on such data, so I am not sure how this
methodology applies or what covets it may hold.

However, R wise, it sounds to me like simply reusing the code here:
http://www.statmethods.net/advstats/factor.html
Would do the trick.

p.s: you wrote "It would be great to know if there are any R functions that
can do this."
Reply: Please run the following code in your R session -

install.packages("fortunes")
require(fortunes)
fortune("R do")


Cheers,
Tal


----------------Contact
Details:-------------------------------------------------------
Contact me: Tal.Galili@gmail.com |  972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
www.r-statistics.com (English)
----------------------------------------------------------------------------------------------




On Fri, Jul 9, 2010 at 1:16 AM, DHIMAN BHADRA
<dhimanbhadra@gmail.com>wrote:
> Dhiman
	[[alternative HTML version deleted]]