[R-sig-ME] Multivariate Regression with Crossed Random Effects
bbolker at gmail.com
Sat Apr 21 20:56:49 CEST 2018
The ways to "cheat" with 2 variables (i.e., convert the data set to
long/melted format; use the index of the original observation as a
grouping variable) should in principle extend to an arbitrary number
of responses. The problem will be that estimating the 25x25
variance-covariance matrix will be difficult. (Classical MANOVA
approaches make the assumption of sphericity:
https://en.wikipedia.org/wiki/Mauchly%27s_sphericity_test ) You could
use a method (MCMCglmm, glmmTMB, lme) that allows you to constrain the
variance-covariance matrix (e.g. compound symmetric), or a Bayesian
method with a prior on the variance-covariance matrix ...
On Fri, Apr 20, 2018 at 1:14 PM, David Sidhu <dsidhu at ucalgary.ca> wrote:
> Hi All!
> Am I correct that it isn’t currently possible to run a mixed effects linear regression with multiple outcome variables (i.e., multivariate regression) in R? I am talking on the order of 25 outcome variables.
> It seems that there may be some ways to “cheat” lme4 into this, but the examples I’ve come across have all been with a pair of outcomes. Is there any way to do this with 25?
> David M. Sidhu, MSc<http://davidmsidhu.com/>
> PhD Candidate
> Department of Psychology
> University of Calgary
> [[alternative HTML version deleted]]
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models