R help - Nov 2009 - PCA with tow response variables

Hi all,

 

I'm new to PCA in R, so this might be a basical thing, but I cannot find
anything on the net about it.

I need to make a PCA plot with two response variables (df$resp1 and df$resp2)
against eight metabolites (df$met1, df$met2, ...) and I don't have a clue
how to do... and I've only used the simplest PCAs before, like this:

 

pcaObj=prcomp(t(df[idx, c(40:47)]))

biplot(pcaObj)

 

Anyone who knows how to do?

 

Best rageds,

 

Joel
 		 	   		  
_________________________________________________________________
Hitta hetaste singlarna på MSN Dejting!
http://dejting.se.msn.com/channel/index.aspx?trackingid=1002952
	[[alternative HTML version deleted]]

After an off-list email exchange, it sounds like the pls package and Partial
Least Squares are appropriate for this analysis.

Kevin Wright


2009/11/4 Joel Fürstenberg-Hägg <joel_furstenberg_hagg@hotmail.com>
>
> Hi all,
>
>
>
> I'm new to PCA in R, so this might be a basical thing, but I cannot
find
> anything on the net about it.
>
> I need to make a PCA plot with two response variables (df$resp1 and
> df$resp2) against eight metabolites (df$met1, df$met2, ...) and I don't
have
> a clue how to do... and I've only used the simplest PCAs before, like
this:
>
>
>
> pcaObj=prcomp(t(df[idx, c(40:47)]))
>
> biplot(pcaObj)
>
>
>
> Anyone who knows how to do?
>
>
>
> Best rageds,
>
>
>
> Joel
>
> _________________________________________________________________
> Hitta hetaste singlarna på MSN Dejting!
> http://dejting.se.msn.com/channel/index.aspx?trackingid=1002952
>        [[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.
>
>
	[[alternative HTML version deleted]]

Hi all,

 

I wonder what the difference is between the functions prcomp and the PCA
plotting method used in example 3 from the fastICA package. They give totally
different plots. The reason for asking is that I've earlier used prcomp, but
now I should do an ICA, and I guess I cannot compare the PCA plot from prcomp
with the ICA plot if the two PCA plots looks different?

 

Does anyone knows anything about this? Maybe there's a different approach
that's better?

 

 

if(require(MASS))

{
   x <- mvrnorm(n = 1000, mu = c(0, 0), Sigma = matrix(c(10, 3, 3, 1), 2, 2))
   x1 <- mvrnorm(n = 1000, mu = c(-1, 2), Sigma = matrix(c(10, 3, 3, 1), 2,
2))
   X <- rbind(x, x1)
   a <- fastICA(X, 2, alg.typ = "deflation", fun =
"logcosh", alpha = 1,
   method = "R", row.norm = FALSE, maxit = 200,
   tol = 0.0001, verbose = TRUE)
   par(mfrow = c(1, 3))
   plot(a$X, main = "Pre-processed data")
   plot(a$X%*%a$K, main = "PCA components")
   plot(a$S, main = "ICA components")
}


PC=prcomp (X, center=T, scale=T)
hcl=hclust(dist(df))
plot(PC$x[,1],PC$x[,2], main="PCA components (prcomp)")

 

 

Best regards,

 

Joel
 		 	   		  
_________________________________________________________________
Nya Windows 7 gör allt lite enklare. Hitta en dator som passar dig!
http://windows.microsoft.com/shop
	[[alternative HTML version deleted]]