glmmTMB
glmmTMB implements REML by treating the fixed effect parameters as 'random variables', i.e. turning on the Laplace approximation machinery; see http://bbolker.github.io/mixedmodels-misc/glmmFAQ.html#reml-for-glmms for starting points in the literature. https://github.com/glmmTMB/glmmTMB/blob/master/glmmTMB/R/glmmTMB.R#L191 On Mon, Oct 29, 2018 at 6:26 AM Maira Fatoretto
<mairafatoretto at gmail.com> wrote:
Hello,
I am using glmmTMb with binomial family. I chose REML=TRUE because I have
three random effects and I would like to know which is the restriction used
in this estimation in the package.
The REML = TRUE have better estimation than ML in my case.
Thank you.
Best regards.
--
*Ma?ra Blumer Fatoretto*
Estat?stica - Universidade Estadual de Campinas- UNICAMP
Mestra em Ci?ncias(Estat?stica e Experimenta??o Agron?mica) - ESALQ/USP -
Piracicaba - SP
Doutoranda em Estat?stica e Experimenta??o Agron?mica
Telefone:(19) 988309481
E-mail: mairafatoretto at gmail.com
[[alternative HTML version deleted]]
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models