Sunday, May 31, 2015

Lazy Sunday- Tidbit of GLMM


If you're getting an error message about "failure to converge" or anything to do with "eigenvalues"- scaling usually makes the model work.

Don't go to excel file to create new z.scores.

Take the variables you want to look at (If I'm using Poisson, I leave the dependent variable alone) and put it in a data.frame

Then data.frame<-scale(data.frame)

I wish I figured this out sooner.


No comments:

Post a Comment