I read "Longitudinal two-level model" in http://rpsychologist.com/r-guide-longitudinal-lme-lmer.
And I'd like to change the Level1 model a little as below.
Level1
Yij=Beta0j+f(Beta0j)*Rij
Rij~N(0,sigma^2)
f is some function.(e.g.beta distribution,gamma distribution)
Please teach me how to do.
I'd like to use nlme or nlmxir in R packages, but if these packages cannot, teach me other solution (I wish free).
Thanks,
Hi Ryosaku,
Like shown in the example (http://rpsychologist.com/r-guide-longitudinal-lme-lmer), you need to have your data in the form of a dataframe in R. If you already have a table in the form of csv or txt or xls, you can directly import them into R (for example https://www.datacamp.com/community/tutorials/r-data-import-tutorial).
Using the example as an illustration,
lme(y ~ time, random = ~ time | subjects, data=data)
y is the response variable
time is the variable you are interested in
subject would be the variable you use to partition your data.
You mentioned "f is some function.(e.g.beta distribution,gamma distribution)". I do not quite get this. Does your response or variable follow a gamma / beta distribution? In any case, you would have to transform the data first before putting them into the mixed effect model.
best, Lin Gen
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R