[R-sig-ME] Posterior covariance matrix of fixed effects from MCMCglmm?

Ben Bolker bbolker at gmail.com
Fri Oct 30 21:00:11 CET 2015


  [please keep r-sig-mixed-models in Cc: list]

  In my mind they are the same thing, but to be sure you'd have to
define exactly what *you* mean by "[the] posterior covariance matrix for
fixed effects" and how you intend to use/interpret the results you get ...

  Ben Bolker

On 15-10-30 03:55 PM, Mikhail Matz wrote:
> Hi Ben - sounds super easy, but would not that not  be covariance
> among sampled parameter values along MCMC chain  rather than their
> covariance in the actual data? (or is it the same thing? I’m not
> sire)
> 
> Mikhail
> 
> On Oct 29, 2015, at 8:03 PM, Ben Bolker <bbolker at gmail.com> wrote:
> 
>> What about
>> 
>> var(model$Sol)
>> 
>> ?
>> 
>> On Tue, Oct 27, 2015 at 9:32 AM, Mikhail Matz <matz at utexas.edu>
>> wrote:
>>> Hopefully an easy question: Is there a way in MCMCglmm to extract
>>> a posterior covariance matrix for fixed effects? Many thanks in
>>> advance - Mikhail 
>>> _______________________________________________ 
>>> R-sig-mixed-models at r-project.org mailing list 
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>



More information about the R-sig-mixed-models mailing list