[R] Three way correspondence analyses?

Michael Friendly friendly at yorku.ca
Thu Aug 4 15:25:13 CEST 2016

You haven't supplied any data, and we can only guess which cca() 
function you are using (ade4::cca, ..., vegan::cca(), yacca::cca), and 
the term 'cca' generally refers to canonical correspondence analysis,
which is not quite the same thing as 'three-way correspondence analysis'.

For three-way tables, there are several variations of standard
correspondence analysis that generalize CA for two-way tables
in reasonable, but different ways.
You may find more joy using the mjca() in the ca package
which provides these alternatives.


On 8/2/2016 3:58 PM, Suparna Mitra wrote:
> Hello R experts,
>    have some data for microbiome, metabolome and cytokine from the same
> sample. Now I want to do a three-way correspondence analyses. From three
> normalised data I was trying,
> #Now CCA
> with two data it works good like:
> Metab.Cytok.Microb.cca <- cca(normMicrobiome_NEC,normCytok_and_ProInf)
>  plot(Metab.Cytok.Microb.cca )
> Metab.Cytok.Microb.cca <- cca(normMicrobiome_NEC,normMetab_NEC)
> plot(Metab.Cytok.Microb.cca )
> But when I tried with three
> Metab.Cytok.Microb.cca <-
> cca(normMicrobiome_NEC,normMetab_NEC,normCytok_and_ProInf)
> plot(Metab.Cytok.Microb.cca )
> But this is not displaying all three variables.
> Sorry, I am very new in this. Can anybody please help me?
> Thanks a lot,
> Mitra
> 	[[alternative HTML version deleted]]

More information about the R-help mailing list