[R] nonparametric mixed effect model

[Luisa Carvalheiro]

Hi,

Usually species richness data have very heterogeneous variance, so
they are better described by a Poisson distribution than by a
Gaussian. you should probably analyse your data with Generalized
Linear Models using function lmer (from package lme4) and specify
family=poisson.

hope this helps

luisa

On Wed, Feb 17, 2010 at 4:09 PM, Fons van der Plas
<fonsvanderplas at gmail.com> wrote:
>
> Hi all,
>
> I'd like to compare species richness patterns of two groups with paired
> observations with an extra continuous fixed effect as a covariate. Normally
> I would use a standard general mixed effect model for such data, with one
> group variable and a continuous variable as fixed effects, plus a random
> effect for the paired observations. However, after making such a model with
> the lme function, I saw that the residuals of my model were not normally
> distributed and transforming them to a normal distribution does not seem to
> be possible. Therefore I need a nonparametric equivalent of the lme function
> to build a nonparametric mixed effect model. Does anyone know how I can do
> this? Any help would be very appreciated!
> --
> View this message in context: http://n4.nabble.com/nonparametric-mixed-effect-model-tp1558728p1558728.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>



-- 
Luisa Carvalheiro, PhD
Southern African Biodiversity Institute, Kirstenbosch Research Center, Claremont
& University of Pretoria
Postal address - SAWC Pbag X3015 Hoedspruit 1380, South Africa
telephone - +27 (0) 790250944
Carvalheiro at sanbi.org
lgcarvalheiro at gmail.com