[R-sig-ME] Residual variance random effect GLMM

Ben Bolker bbolker at gmail.com
Tue Jun 21 18:26:52 CEST 2016

  I'll second Thierry's comments, but also point out that there are some
recipes for generalizing the idea of "variance explained": see papers by
Nakagawa and Schielzeth or


  I've now created a tinyURL link for the GLMM FAQ:
http://tinyurl.com/glmmFAQ .  Comments, pull requests, etc. welcome!

     Ben Bolker

On 16-06-20 03:50 AM, Thierry Onkelinx wrote:
> Dear Sara,
> Unlike a linear model, generalised linear models don't have a residual
> variance. A linear model assumes a Gaussian distribution with two
> parameters: mean and standard error which are independent. Generalised
> linear models use distributions how dependent on only one parameter
> (binomial, Poisson). Mean and variance of those distributions are defined
> by the same parameter. In case a generalised linear model uses a two
> parameter distribution (e.g. negative binomial), still the mean and
> variance are influenced by a common parameter (mean = mu, var = mu + mu ^ 2
> /theta).
> Best regards,
> ir. Thierry Onkelinx
> Instituut voor natuur- en bosonderzoek / Research Institute for Nature and
> Forest
> team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
> Kliniekstraat 25
> 1070 Anderlecht
> Belgium
> To call in the statistician after the experiment is done may be no more
> than asking him to perform a post-mortem examination: he may be able to say
> what the experiment died of. ~ Sir Ronald Aylmer Fisher
> The plural of anecdote is not data. ~ Roger Brinner
> The combination of some data and an aching desire for an answer does not
> ensure that a reasonable answer can be extracted from a given body of data.
> ~ John Tukey
> 2016-06-18 17:30 GMT+02:00 Fraixedas, Sara <sara.fraixedas at helsinki.fi>:
>> Dear all,
>> I want to calculate the percentage variance explained by a random effect
>> in a GLMM fitted using the "glmmADMB" package. For that I would need to
>> know what is the residual variance but it is not given in
>> "VarCorr.glmmadmb" or in the summary output command.
>> How can I extract the residual variance from a random effect in this
>> particular case?
>> Thank you in advance,
>> Sara Fraixedas
>> Doctoral Student
>> The Helsinki Lab of Ornithology (HelLO) Finnish Museum of Natural
>> History P.O. Box 17
>> 00014 University of Helsinki, Finland
>> Tel. +358-9-19128851
>>         [[alternative HTML version deleted]]
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> 	[[alternative HTML version deleted]]
> _______________________________________________
> 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