Hi,
It seems like the package "mvpart" is quite outdated and not available
for
the current R release. Since PCA is a very common need I'll suggest finding
a replacement for it so that the error will either go away, or it is easier
for us to reproduce it.
Best,
Jiefei
On Tue, Aug 25, 2020 at 1:06 PM Andrew Halford <andrew.halford at
gmail.com>
wrote:
> Hi Listers
>
> Using mvpart to run a MV regression tree with PCA= TRUE to get a PCA
> plotted with sites coloured according to the tree output.
>
> Unfortunately it wont produce the PCA, instead giving the error message..
>
> Error in cor(xall, xx[order(tree$where), ]) : incompatible dimensions.
>
> However, when I run a PCA on the data using the rda command I have no
> problems producing a PCA.
>
> data is attached as a text file
>
> my code is thus...
> fish05.hel <- decostand(fish05,"hellinger")
> fish05.mrt <-
>
>
mvpart(data.matrix(fish05.hel)~.,env,margin=0.08,cp=0,rsq=TRUE,xv="pick",xval=nrow(fish05),xvmult=100,which=4,pca=TRUE)
>
> The tree is produced no problem but it wont produce a PCA.
>
> I am just keen to understand what this error means as I dont see anything
> unusual about the dataset used, notwithstanding the data is rather messy.
>
> Andy
>
>
>
>
>
> --
> Andrew Halford Ph.D
> Senior Coastal Fisheries Scientist
> Pacific Community | Communaut? du Pacifique CPS ? B.P. D5 | 98848 Noumea,
> New Caledonia | Noum?a, Nouvelle-Cal?donie
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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]]