Covariance explained by PLS axes
Closed this issue · 2 comments
This is a suggestion. The function 'phylo.integration' and 'two.b.pls' could return the percentage of the covariance explained by each of the PLS pair of axes. This would be an interesting feature to explore how many axes one need to explain most of the covariation in the data.
Actually, the two.b.pls function provides that information, as one of the objects output in the list is the SVD of the cross-covariance matrix. But it's a good suggestion to do the same in integration.test and phylo.integration. Thank you!
I'm glad this comment has been made, because I missed the value expressing percentage covariance explained by phylo.integration. Especially because this allows a great comparison of explanatory power of phylogenetic and non-phylogenetic 2B-PLS.