Hi everyone,
I have a problem with the compar.gee function when ruuning a phylogenetic logistic regression with the following error message:
Error in gee(binary.y ~ X2.1 + binary.X3, c(1, 1, 1, 1, 1, 1, 1, 1, 1, :
Cgee: error: logistic model for probability has fitted value very close to 1.
estimates diverging; iteration terminated.
Cgee: error: logistic model for probability has fitted value very close to 1.
estimates diverging; iteration terminated.
This issue has already been discussed between Emmanuel Paradis and Kaspar Delhey. Emmanuel argued that "The error comes certainly from that species with "ones" are very closely related (or a similar pattern) making impossible to estimate correctly the binomial parameters. You may try transforming the branch lengths of the tree to check this." And Kaspar then said it worked when he transformed the branch length with e.g. tree2=compute.brlen(phy=tree1, method="Grafen", power=0.5) and also it would have worked when he uses a tree with all branch lenghts equal to 1. However, I tried those things and it didn`t work in my case and more importantly, it does not make sense to me to transform the branch lengths `cos the structure of the tree does not change, only the branch lengths change, but all the 1`s are still in the closely related species. Can anyone explain that to me why transforming branch lengths would change anything? How can I evade this error?
For any help I am very grateful.
Best,
Sereina
_______________________________________________ R-sig-phylo mailing list - R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/
