I am interested in running a model predicting a continuous variable, using several fixed effects as well as random subject and item effects. I would like to estimate the model using MCMC sampling (or some equivalent method). I'm not interested in fitting priors. What is the easier way to do this in R?
For background you might like to read the MCMC chapter in this
Book Developing multilevel models for analysing contextuality, he...
and the resources here
http://www.bristol.ac.uk/cmm/software/mlwin/mlwin-resources.html#estimation
while
http://www.bristol.ac.uk/cmm/software/r2mlwin/
allows you to use MLwiN from within R
Thank you Kelvyn Jones for sharing your knowledge. Hope this will be helpful for us too.
I would recommend the R package MCMCglmm by Jarod Hadfield. There are quite extensive vignettes for this package on CRAN and Jarod is very active on the R-sig mixed models mailing list. And there are some good examples comparing it with alternative mixed modelling packages eg https://groups.nceas.ucsb.edu/non-linear-modeling/projects/owls/WRITEUP/owls.pdf
Try brms (my recommendation) and rstanarm packages, which allow you to use a syntax similar to the one used with the lme4 package, but from a Bayesian perspective (MCMC using Stan -Hamiltonian Monte Carlo-). Alternatively, use the inla (integrated nested Laplace approximations) package as an alternative to MCMC techniques in Bayesian inference. In any case, be careful with the default priors: they can be a pain.
Regards.
- https://cran.r-project.org/web/packages/brms/index.html
- https://cran.rstudio.com/web/packages/rstanarm/
- http://www.r-inla.org/
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R