Hi Kaspar, Matt, Marguerite and others,
I provide an overview of the issue of intraspecific variation in
comparative methods in the 2nd edtion of my book. Specifically, there
are functions pic.ortho and varCompPhylip (the latter requiring Phylip
on your machine) which implement most of the met
Dear all,
I used a very raw 18S phylogeny of Craniates to introduce phylogeny to
students, and I would like to estimate bootstrap or likelihood values
for branches. Sorry for this long message, but I hope you can answer at
least some of my questions.
1) For bootstrapping, I obtain "zero" valu
Dear Tom,
There is no reason to assume that the residuals from your two PGLS analyses
will be independent of phylogeny, so if you are going to do this you should
correlate the residuals phylogenetically (i.e. run them through PGLS).
General problems with using residuals as data have been commente
Dear Gwennaƫl,
first it seems there was a bug in the function prop.clades from ape
introduced recently, results in many zeros of bootstrap values.
This function needs to be replaced in the ape package:
prop.clades <- function (phy, ..., part = NULL, rooted = FALSE)
{
if (is.null(part)) {
I have a doubt with using the picante::phylosor.rnd for testing hypothesis
about the phylogenetic dissimilarities between sites. This function gives
the randomized matrices obtained with the null model chosen. But how can I
do to explicitly test for a significant difference between the observed
phy
Hi all
I have a dataset of 5 PC axes for a phylogeny with 52 tips and I would like
to get the variance-covariance matrix under a global OU model of
evolution. I find that OU is strongly favored over BM, based on
fitContinuous in GEIGER.
However, I am having issues with convergence when trying to f
You might want to try the OUWie package by Beaulieu & O'Meara
(http://cran.r-project.org/web/packages/OUwie/index.html). I have not
tried it yet, but it promises to do multi-optimum OU model fitting as well.
All the best, Liam
--
Liam J. Revell
University of Massachusetts Boston
web: http://fa
Thanks Rob and Alejandro,
OK, I did as suggested and ran a PGLS with A ~ B + C. I was hoping for some
clarification of the actual results. Here is a summary:
**
Generalized least squares fit by maximum likelihood
Model: variableA ~
Thanks for the response, Liam. I just tried out OUwie and it works.
However, it appears to only fit 1 variable at a time (I only get one sigma
squared value).
When running some tests with OUCH, I found that sigma squared is different
when one variable is fit to an OU model, versus when multiple var
Just wanted to note that there's a fundamental difference between the
way OUCH is handling the multivariate traits and the way geiger and
similar packages are handling it. If you're giving geiger a data
frame with multiple traits, it is fitting each independently, just
saving you the trouble from
That helps alot, Carl, thank you. So to bring this back to my original
question, it sounds like the only way to get sigma squared values from a
multivariate OU model is with OUCH. I'm currently not able to get results
because of convergence issues. Does that mean that I simply won't be able
to get
Hi Pascal,
Yes, there's lots of ways to fiddle with the optimization routines in
OUCH. The first thing would be to try more iterations. You could
also switch the optim routine (it defaults to Nelder Mead, you could
be like geiger and use L-BFGS etc).
For example, with more iterations:
hansen(da
12 matches
Mail list logo