://scholar.google.com/citations?user=nBSC4-EJ=it
- Original Message -
From: "Jarrod Hadfield"
To: "David COSTANTINI"
Cc: "r-sig-phylo"
Sent: Monday, 17 May, 2021 21:34:34
Subject: Re: [R-sig-phylo] phylogenetic correction and MCMC model
Hi,
chronos adds a chronos
://davidcostantini.wordpress.com/
https://twitter.com/DavidZool
http://scholar.google.com/citations?user=nBSC4-EJ=it
- Original Message -
From: "Jarrod Hadfield"
To: "David COSTANTINI"
Cc: "r-sig-phylo"
Sent: Monday, 17 May, 2021 21:34:34
Subject: Re: [R-sig-phylo] phylogen
Ecology
http://davidcostantini.wordpress.com/
https://twitter.com/DavidZool
http://scholar.google.com/citations?user=nBSC4-EAAAAJ=it
- Original Message -
From: "Jarrod Hadfield"
To: "David COSTANTINI" , "r-sig-phylo"
Sent: Monday, 17 May, 2021 19:48:59
Subject: R
Hi David,
It looks like phylo_ultra might be a list? Is phylo_ultra[[1]] a tree?
Also, don't use nodes="TIPS"; this is just to demonstrate how poor the
algorithm is when you don't use the expanded inverse. I see people using
nodes="TIPS" a lot - where does this code come from?
Cheers,
Jarrod
Hi Liam,
In multi-level models DIC can be 'focused' at different levels. In
MCMCglmm, DIC is focussed at the highest possible level because this is
the only level at which it can be analytically computed for non-Gaussian
models. The highest level is not the level at which most scientists want
Hi Jesse,
In order to account for phylogenetic uncertainty you are better just
pulling trees from their posterior rather than choosing trees that are
incongruent. The latter will give estimates biased toward those
associated with extreme trees.
If the analysis is the binomial model you
Dear Jarrod,
Thanks very much for the quick reply.
I'll try to implement the changes in the model.
Have a nice weekend,
Diogo
Em Sex, 14 de jul de 2017 17:48, Jarrod Hadfield
<j.hadfi...@ed.ac.uk <mailto:j.hadfi...@ed.ac.uk>> escreveu:
Hi Diogo,
Hi Diogo,
First, your model1 is unlikely to be valid unless the residual variance
happens to be 1. You should not fix it at one, and use a prior like:
prior = list(R = list(V = 1, nu = 0.02), G=list(G1=list(V=1, nu=0.02)))
Note that the residual variance (Ve) is the intra-specific variance,
data=traits,
prior=prior.dep2,
pr=TRUE,
pl=TRUE,
family="threshold")
an send me the summary and hist(dep2$Liab)
Cheers,
Jarrod
On 16/12/2016
l=TRUE,
family="threshold")
an send me the summary and hist(dep2$Liab)
Cheers,
Jarrod
On 16/12/2016 07:02, Jarrod Hadfield wrote:
Hi Chris,
I think ngen in threshbayes is not the number of full iterations (i.e.
a full update of all parameters), but the number of full iterat
Hi Chris,
I think ngen in threshbayes is not the number of full iterations (i.e. a
full update of all parameters), but the number of full iterations
multiplied by the number of nodes (2n-1). With n=600 species this means
threshbayes has only really done about 8,000 iterations (i.e. about
derive from your specification of the priors. Usually you don’t
specify the prior for B in MCMCglmm. The problem may also be related to the
size of your dataset. Estimation of effects can be difficult with binary data,
when the dataset is small. Below is a small example from Jarrod Hadfield for
b
Dear Diederik,
The lack of convergence is because the residual variance is
non-identifiable with binary data but you have a very weak prior on
it. You should fix the residual variance at something (I usually use 1):
prior.test<-list(R=list(V=1,fix=1), G=list(G1=list(V=1, nu=0.002),G2 =
for the question. This
should be easier for a continuous predictor, right?
Cheers
Gustaf
On 2015-01-22 12:23, Jarrod Hadfield wrote:
Hi Gustaf,
In the model with just species the residual variation is
measurement error/plasticity error, but could also include
deviations from the assumed BM process
we
observe:
Va / (Va+Vhab+Ve) #phylo
Vhab / (Va+Vhab+Ve) #habitat
Ve / (Va+Vhab+Ve) #measurement/plasticity/local adaption and
other processes
Did I get that right or am I lost?
Gustaf
On 2015-01-22 04:54, Jarrod Hadfield wrote:
Hi Gustaf,
1/ You can ignore nhabitat: for some reason
Dear Gustaf,
How many levels of `habitat' are there, and are they cross-classified
with respect to species (i.e. are multiple species measured in the
same habitat)?
Assuming for now there are a reasonable number of habitats then the
simplest model (without cross-classification) in
2013 um 14:54 Uhr
VON: Jarrod Hadfield j.hadfi...@ed.ac.uk
AN: Sereina Graber sereina.gra...@gmx.ch
CC: r-sig-phylo@r-project.org
BETREFF: Re: Aw: Re: [R-sig-phylo] WG: Re: Re: MCMCglmm for
categorical data with more than 2 levels - prior specification?
Hi,
They are the effect of the covariates
Hi Sereina,
You should not get that error message when you do not specify a prior
- but if you do can you let me know.
For the prior you specified you get the error message because
us(trait):units is specifying a 3x3 covariance matrix, and yet your
prior, R=list(V=1,nu=0.002), is
Hi Sam,
The terminology G and R structure is used widely, for example in
ASreml SAS and probably others. The G-structure is the covariance
matrix of the random effects and the R-structure is the covariance
matrix of the residuals. In your model you have one random term
(animal) and one
Hi,
ASReml is another option, which uses REML. It takes 1/10th of a second
on a 1000 tip phylogeny and is considerably more flexible.
fit-asreml(y~x,random=~giv(species),data=dat,ginverse=list(species=sm2asreml(Ainv)))
# with the data set up as:
ntips-1000
tree-rcoal(ntips) #
Hi,
Quoting Margaret Evans mekev...@yahoo.com on Mon, 24 Sep 2012
22:56:48 +0100 (BST):
Hello all,
I have a few questions concerning the specification of flat priors
(on the probability scale) for a phylogenetic logistic regression in
MCMCglmm.
1) First, I'd like to verify my
Hi,
Regarding the blog and the feasibility of MCMCglmm for threshold models:
If y1 is binary and y2 is normal, then the univariate analysis would be:
Ainv-inverseA(tree)$Ainv
m1-MCMCglmm(y1~y2, random=~species,ginverse=list(species=Ainv),
data=my.data, prior=my.prior, family=ordinal)
Hi,
Thanks for the Allman Rhodes paper, it is very nice. For me at least
it confirms my suspicions, but made me realise that claims of
asymmetric transition rates are only suspicious if you are unprepared
to make some (strong?) assumptions. If anyone disagrees with what I
have written
On Aug 17, 2012, at 6:31 AM, Jarrod Hadfield wrote:
Hi,
Thanks for the Allman Rhodes paper, it is very nice. For me at
least it confirms my suspicions, but made me realise that claims
of asymmetric transition rates are only suspicious if you are
unprepared to make some (strong?) assumptions
Hi,
I have been helping someone with some analyses and came across some
routines to estimate asymmetric transition rates between discrete
characters. This surprised me because its fairly straightforward to
prove that asymmetric transition rates cannot be identified using data
collected
supported
y-rbinom(n, 1, 0.5) # random data unconnected to the tree but p=0.5
m1-ace(y, tree, type = d, model=SYM)
m2-ace(y, tree, type = d, model=ARD)
anova(m1, m2) # asymmetric evolutionary transition not supported
Cheers,
Jarrod
Quoting Jarrod Hadfield j.hadfi...@ed.ac.uk on Thu, 16
On Aug 16, 2012, at 10:09 AM, Jarrod Hadfield wrote:
Hi,
I have had a few replies off-list which have made me try and clarify
what I mean. I think the distinction needs to be made between two types of
probability: the probability that an outcome is 0 or 1 Pr(y| \theta) and
the probability density
Dear Wayne,
This is my fault. With phylogenies the ancestral nodes are treated as
missing data and so I set their measurement error to an arbitrary
value. The code for working out how many new measurement errors
there are was incorrect.
L98 of MCMCglmm.R should read
mev-c(mev, rep(1,
28 matches
Mail list logo
