Mixed effects modelling

Mixed effects modelling


Last edit: Kiyosada Kawai (13 Dec 2016 14:45) | Revisions: 1 | Created by: Kiyosada Kawai | Rating: 0


Generalized linear mixed model (GLMM)

1. From GLM to GLMM

Generalized linear model (GLM) is useful when estimating parameters.
GLM consists of three parts.

  • Probability distribution (e.g., Poisson, normal)
  • Link function (e.g., logit, log)
  • Linear predictor

However, GLM is not enough to deal with the observed data
because the data usually contain much higher variance (overdispersion) than GLM predicts.
Source of variation comes from like differences of individuals or locations, which GLM
does not take account.

2. GLMM in R

There are several ways to do GLMM analysis in R….

GlmmML
glmer
lmer
lme

3. Reference

Kubo 2008
http://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/49477/7/kubostat2008f.pdf


rating: 0+x
Add a New Comment
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License