Hello r-helpers I have a correlation matrix (either pearson or spearman) of 41 variables and i want to use it to construct a tree with agnes command (instead of using the distances). My problem is that at the height of the tree i see negative scaling also which is not desired. Is there any r-command that uses as a distance method the correlation estimates instead of distances (distances=dissimilarities, correlations=simmilarities) so that the height of the tree will be between 0 (zero) and 1 (one). Or simply, is there a way to transform the correlations in a proper way and to use the agnes command for example so that the height of the tree will be between 0 and 1?Thank you very much. _____________________________________________________________________________________ http://domains.europlanet.gr ???? ?????????? ?????? ?????? internet ???????? ???? 10 ????????.
On 12/30/05 5:27 AM, "vagua1 at mailbox.gr" <vagua1 at mailbox.gr> wrote:> Hello r-helpers > > I have a correlation matrix (either pearson or spearman) of 41 variables and i > want to use it to construct a tree with agnes command (instead of using the > distances). My problem is that at the height of the tree i see negative > scaling also which is not desired. Is there any r-command that uses as a > distance method the correlation estimates instead of distances > (distances=dissimilarities, correlations=simmilarities) so that the height of > the tree will be between 0 (zero) and 1 (one). Or simply, is there a way to > transform the correlations in a proper way and to use the agnes command for > example so that the height of the tree will be between 0 and 1?Thank you very > much.1-correlation? Sean