[R-sig-ME] multilevel circular regression

[Henrik Parn]

Dear all,

I wish to analyse the dispersal direction (angle 0-360) in juvenile 
birds in relation to e.g. their sex. However, the movement may be 
non-independent among individuals. For example juveniles from the same 
nest may have non-independent dispersal behaviour due to genetic or 
environmental effects, or individuals from nests in the same colony may 
have a similar dispersal directions. Therefore, I wish to analyse 
individual dispersal direction in a nested model. Something like:

direction ~ sex + (1|colony.id) + (1|nest.id:colony.id)
...using lmer-ish notation.

I have seen similar questions on the forum before
http://finzi.psych.upenn.edu/Rhelp10/2009-May/198498.html
http://finzi.psych.upenn.edu/R/Rhelp02/archive/53700.html
...but could not find any answers

I am aware of the packages circular and CircStats, and in particular the 
function lm.circular.cl for a non-nested regression.

Has anyone tried mixed model circular regression using R? Any hints of 
relevant packages or functions are welcome!

Thanks a lot in advance!

Henrik


-- 
Henrik Pärn
Centre for Conservation Biology
Department of Biology
Norwegian University of Science and Technology
NO-7491 Trondheim
Norway

Office: +47 73596285
Fax: +47 73596100
Mobile: +47 90989255

E-mail: henrik.parn at bio.ntnu.no