[adegenet-forum] spatial PLS - a query

Ondřej Mikula onmikula at gmail.com
Tue Aug 21 14:24:40 CEST 2012


Thank you for your prompt answer.
I was inspired by one particular variant of PLS which is achieved
simply by singular value decomposition of
1/N * t(X) %*% Y
Looking at this equation I am wondering if it is possible to use
straightforward analogy
svd( 1/(2*N) * t(X) %*% (W + t(W)) %*% Y)
to construct spatial PLS. But perhaps it is not so easy to answer such
question with certainty.
Best
Ondřej


On 21 August 2012 13:57, Jombart, Thibaut <t.jombart at imperial.ac.uk> wrote:
> Hello,
>
> That is an interesting question, but having never used PLS approaches, I don't know.
>
> Cheers
>
> Thibaut
> ________________________________________
> From: adegenet-forum-bounces at lists.r-forge.r-project.org [adegenet-forum-bounces at lists.r-forge.r-project.org] on behalf of Ondřej Mikula [onmikula at gmail.com]
> Sent: 21 August 2012 12:44
> To: adegenet-forum at lists.r-forge.r-project.org
> Subject: [adegenet-forum] spatial PLS - a query
>
> Dear Thibaut,
> I have one theoretical question. Is it possible to generalize spatial
> PCA to spatial partial least squares? I wished to look for spatially
> structured covariances between two sets of variables.
> Best wishes
> Ondřej
>
> --
> Ondřej Mikula
>
> Institute of Animal Physiology and Genetics
> Academy of Sciences of the Czech Republic
> Veveri 97, 60200 Brno, Czech Republic
>
> Institute of Vertebrate Biology
> Academy of Sciences of the Czech Republic
> Studenec 122, 67502 Konesin, Czech Republic
> _______________________________________________
> adegenet-forum mailing list
> adegenet-forum at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum



-- 
Ondřej Mikula

Institute of Animal Physiology and Genetics
Academy of Sciences of the Czech Republic
Veveri 97, 60200 Brno, Czech Republic

Institute of Vertebrate Biology
Academy of Sciences of the Czech Republic
Studenec 122, 67502 Konesin, Czech Republic


More information about the adegenet-forum mailing list