[R-sig-phylo] How to incorporate intraspecific variation in MCMCglmm

[Diogo B. Provete]

Dear list,
I'm trying to fit a mixed-effects model to a problem that includes:
a continuous response variable (mean jumped distance by 17 species of
frogs) in 3 different arenas and under 2 types of stimuli. So, it's a kind
of two-way ANOVA design. I have the species phylogeny and prepared its
inverse:

treeAinv<-inverseA(phylo,nodes="TIPS",scale=TRUE)$Ainv

I included the following priors:

prior = list(R = list(V = 1, fix = 1), G=list(G1=list(V=1, nu=0.02)))

and the model:

model1<-MCMCglmm(mean_distance~type_arena*type_of_stimulus,
 random=~Specie, data=df_spe, family="gaussian",
 ginverse = list(Specie=treeAinv), nodes="ALL",
prior=prior, nitt=30, burnin=25000, thin = 100, verbose=FALSE)

which seems to converge well (heidel.diag p-value = 0.428), ESS greater
than 1000 for both random and fixed effects.

My question is: how can I incorporate intraspecific variation in the jumped
distance? I have the standard error of the distance and tried to make the
model as:

random=~Species+ us(1+se_distance):Species

but I keep getting an error about prior specification:

Error in MCMCglmm(mean_distance ~ type_arena * type_og_stimulus, random =
~Specie +  :
  prior$G has the wrong number of structures

Thank you in advance,
Diogo




-- 
*Diogo B. Provete, PhD.*
FAPESP Post-doctoral fellow
Department of Environmental Sciences
Centre for Sciences and Technologies for Sustainability
Federal University of São Carlos
Sorocaba

Gothenburg Global Biodiversity Centre
Box 462   SE-405 30
Göteborg, Sverige

diogoprovete.weebly.com

Cell phone: +5515981022137
Skype: diogoprovete

Editor: Amphibia-Reptilia | Biodiversity Data Journal

[[alternative HTML version deleted]]

___
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/

Re: [R-sig-phylo] How to incorporate intraspecific variation in MCMCglmm

[Jarrod Hadfield]

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, 
assumed constant over species, as in Lynch's h2/Pagels Lambda. If you 
have the standard error (se) of each observation you can also allow 
heterogeneity between observations as in random-effect meta-analysis:


random=~Specie+idh(se)

If you fix the variance associated with se at 1 in the prior

prior = list(R = list(V = 1, nu = 0.02), G=list(G1=list(V=1, nu=0.02), 
G2=list(V=1, fix=1)))

Then the intraspecific variance for species i is se^2_i+Ve

If you do not fix the variance associated with se at 1, for example with

prior = list(R = list(V = 1, nu = 0.02), G=list(G1=list(V=1, nu=0.02), 
G2=list(V=1, nu=0.02)))

and estimate the associated variance (Vse). Then the intraspecific variance
 for species i is Vse*se^2_i+Ve. This is useful if you only know the standard 
error up to proportionality.

Alternatively you can fit fixed-effect meta-analysis where all the 
intraspecific variance is presumed to be due to sampling error:

random=~Specie, rcov=~idh(units):units

with prior:

prior = list(R = list(V = diag(se^2), fix=1), G=list(G1=list(V=1, nu=0.02)))

The intraspecific variance for species i is se^2_i. This is quite an 
inefficient way of doing fixed-effect meta-analysis, and one day I will make 
such analyses easier to fit, and quicker to fit..

Also, don't use nodes="TIPS" in the call to inverseA. I only allowed this option because 
its the parameterisation used by most other software, but its a really bad way of doing it. Stick 
with the default, node="ALL".

Cheers,

Jarrod

 








On 14/07/2017 14:26, Diogo B. Provete wrote:

treeAinv<-inverseA(phylo,nodes="TIPS",scale=TRUE)$Ainv

I included the following priors:

prior = list(R = list(V = 1, fix = 1), G=list(G1=list(V=1, nu=0.02)))



--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

___
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/