[R-sig-ME] Help with General Additive Mixed Model

[Leisten, L. (Luca)]

Dear all,


I have a little bit of an off topic question regarding general additive mixed models (GAMM).


I am using the bam (or gam) function of the mgcv package in R and have some model specification and interpretation issues. As I have no experience with GAM models I was wondering if there is anyone around that has experience with them and could help me answer some questions?


Specifically my questions concern (1) the model specification of my model, as I get different results compared to a linear mixed model that includes poly terms, (2) the differences between the gam and bam function, as both give me different estimates, (3) and the difference between running a full model vs running two subsets as again, those give me different conclusions.


I would very much appreciate any help and would love a meeting if anyone has some time available to help! I'm not sure how the email list works, but in case you'd like to contact me directly my email is Luca.leisten using ru.nl.


Thanks for your help!


Best,

Luca Marie Leisten


	[[alternative HTML version deleted]]