nested mixed effects logistic regression binomial, glm) results differ by function.
---------------------------------------------------------------------- Message: 1 Date: Thu, 23 Apr 2015 18:16:26 +0200 From: Lize van der Merwe <lizestats at gmail.com> To: <r-sig-mixed-models at r-project.org> Subject: [R-sig-ME] nested mixed effects logistic regression binomial glm) results differ by function. Message-ID: <009401d07de0$dba71190$92f534b0$@gmail.com> Content-Type: text/plain; charset="UTF-8" Please advise: I have a dichotomous outcome on 2500 individuals. From 18 geographical areas, and many households nested within areas. I need to assess the association between various predictors and my outcome, adjusting for the correlation within households, as well as within areas. The following R functions provide dramatically different results. glmer(CC~predictor+1|area/household,family=binomial) and glmmPQL(CC~predictor, random=~1|area/household),family=binomial) Why? Which is correct?
Instead of focusing on the 'which one is correct' question, I would focus more on 'why are they different'. If two techniques give you rather different results then perhaps your model is too complicated, or your data set does not allow for a 2-way nested model. Simulate a data set with 2500 observations with nicely balanced clusters and see whether you have the same problems. Then cripple the data set (make unbalanced clusters) and see when differences occur. And yes....doing it in a Bayesian context is a good idea too. Alain
Thanks in advance. (I posted this on another site too.) Lize
Dr. Alain F. Zuur First author of: 1. Beginner's Guide to GAMM with R (2014). 2. Beginner's Guide to GLM and GLMM with R (2013). 3. Beginner's Guide to GAM with R (2012). 4. Zero Inflated Models and GLMM with R (2012). 5. A Beginner's Guide to R (2009). 6. Mixed effects models and extensions in ecology with R (2009). 7. Analysing Ecological Data (2007). Highland Statistics Ltd. 9 St Clair Wynd UK - AB41 6DZ Newburgh Tel: 0044 1358 788177 Email: highstat at highstat.com URL: www.highstat.com