I'd like to analyse asymptotic data using the nlme package in R but cannot figure out how to specify two crossed random effects (instead of nested random effects).
The data comes from a computational modelling study. The dependent variable is test accuracy as predicted by time (iteration) and the random intercepts should be network and trial. I'm using the following code for the base model:
base_model
The R script below illustrates the nested versus non-nested (crossed) random effects functionality in the R packages lme4 and nlme. Note that crossed random effects are difficult to specify in the nlme framework. Thus, I've included a back-of-the-envelope (literally a scanned image of my scribble) interpretation of the 'trick' to specifying crossed random effects for nlme functions (i.e., nlme and lme).
## This script illustrates the nested versus non-nested
## random effects functionality in the R packages lme4 (lmer)
## and nlme (lme).
library("lme4")
library("nlme")
data("Oxide")
Oxide
Thanks a lot for your explanation!
I actually tried the pdBlocked and pdIdent functions and R gives the following error:
Error in nlme.formula(Accuracy ~ SSlogis(Iter, Asym, xmid, : 'random' must be a formula or list of formulae
Do you have any thoughts on that? I think I used the exact same code as you described above but maybe I'm wrong and missed something.
Thanks again!
It's not easy to specify crossed random effects with nlme. If possible, use lme4 instead.
See: https://stats.stackexchange.com/questions/362334/r-2-way-repeated-measures-anova-different-degree-of-freedom-among-ezanova-a/
Here is a highly recommended answer in StackExchange FYI,
https://stats.stackexchange.com/questions/228800/crossed-vs-nested-random-effects-how-do-they-differ-and-how-are-they-specified
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics