1

I am using glmmPQL (from nlme package) to run a multi-level glm with a dummy outcome. Unfortunately glmmPQL does not provide fit measures such as AIC, or BIC. I was wondering if there is a package that can calculate AIC, and BIC fromglmmPQL results.

UseR2001
  • 29
  • 4
  • this may help answer the question: http://anson.ucdavis.edu/~mueller/SJS-02-117-Chiou.pdf – Dij Apr 15 '19 at 04:16
  • _"Since glmmPQL uses quasi-likelihood, technically there is no log-likelihood (hence no AIC nor BIC, which are based on the log-likelihood) for this model"_ (*@zhijie zhang* 2008, found here: https://stat.ethz.ch/pipermail/r-help/2008-December/181937.html) – jay.sf Apr 15 '19 at 05:17
  • Thank you, I also found this which talks about Quasi AIC-BIC: https://mirror.its.sfu.ca/mirror/CRAN/web/packages/bbmle/vignettes/quasi.pdf – UseR2001 Apr 15 '19 at 22:02

0 Answers0