[R-sig-ME] lmer() with no intercept (array chip)

Elizabeth Crone elizabeth.crone at cfc.umt.edu
Sat Jun 5 18:34:53 CEST 2010

> Hi, I am wondering how I can specify no intercept in a mixed model using lmer().

The intercept IS the term that varies randomly in both your models, i.e., (1|worker) and (1|day).  If you look at the terms you get, e.g., from coef() and ranef(), you should find that you are getting the same "intercept" terms for each group in the "no intercept" and "intercept" models, they are just parameterized differently.

If you want to play around with random effects, no intercept models, you need to make something other than the intercept vary among groups, e.g.,
y~x+x2+(x+x2|worker)+(x+x2|day) vs. y~-1 + x+x2+(-1+x+x2|worker)+(-1+x+x2|day)

Good luck!

More information about the R-sig-mixed-models mailing list