m of the shared branch lengths. Is
this appropriate? Why or why not?
Any input would be much appreciated.
Cheers,
Peter Smits
On Mon, Aug 11, 2014 at 3:34 PM, Edwin Lebrija Trejos
wrote:
>
> Dear all,
>
>
> I am plant ecologist using a 2-level hierarchical Bayesian model
> (inpleme
scaling of branch lengths, I agree with Joe that there is
> nothing particular about 1, other than providing an easier interpretation
> for the numerical value of the phylogenetic variance.
> Cheers,
> Cecile.
>
>
> On 12/14/2014 01:03 PM, Peter Smits wrote:
>
>>
Hi Will,
Quick answer to that question: yes.
The key is that categorical variables cannot be modeled directly in a GLM
framework. These categorical variables, or index variables, are transformed
into n x (k - 1) matrices of index variables. These index variables are
binary where a 1 corresponds t
Quick correction: index variables are transformed into n x (k - 1) matrices
of INDICATOR variables.
On Wed, Feb 4, 2015 at 5:25 PM, Peter Smits wrote:
> Hi Will,
>
> Quick answer to that question: yes.
>
> The key is that categorical variables cannot be modeled directly in a
Hi Andrea,
To paraphrase Gelman and Hill 2007 on regression modeling: a coefficient
estimate is considered significant when the mean/modal estimate is more
than 2 standard errors away from 0. This means, your beta estimate
(e.g. 0.1292656)
is known with some kind of error. If the estimate - 2*erro
The alternative to MCMCglmm would be to use stan or bugs for writing your
own sampling statement + priors. You'll have more control than with
MCMCglmm, but it will have even more of a learning curve.
Using stan will also most likely be faster than using any single R package.
Cheers,
Peter
On We
Hi,
If your adventurous you could probably write your own model in R or Stan
or similar. The wrapped Normal or the Von Mise distribution are circular
and map from 0 to 2pi. My suggestion would be to include a species level
phylogenetic "random effect" for the mean parameter to account for the
aut
