[R-sig-ME] gamm4: predict to reflect random effects?

Mark Miller mark.gr.miller at gmail.com
Mon Oct 13 08:40:14 CEST 2014

Hi Mollie,

Thanks for the reply, yep I guess for now using gam is the best option of
simple random effects. I am trying to fit a more complex model with nested
random effects and want both slope and intercept to vary -  I'll try using
splines within a GLMM and seeing how that works out!



On Fri, Oct 10, 2014 at 9:30 PM, Mollie Brooks <mbrooks at ufl.edu> wrote:

> Hi Mark,
> Since no one replied to my question, I assume that the answer given by
> Simon Wood in 2011 is still current.
> http://r.789695.n4.nabble.com/mgcv-gamm-predict-to-reflect-random-s-effects-td3622738.html
> As Wood suggests at the link (scroll down), my lab group is using gam()
> with a spline for the random effect like s(Subject,bs="re")
> Cheers,
> Mollie
> ------------------------
> Mollie Brooks, PhD
> Postdoctoral Researcher, Population Ecology Research Group
> Institute of Evolutionary Biology & Environmental Studies, University of
> Zürich
> http://www.popecol.org/team/mollie-brooks/

	[[alternative HTML version deleted]]

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