Monday, Febuary 27, 2012 14h-16h (N4/17 - Stewart Biology)Zofia Ecaterina Taranu## Topics- How to address non-independence among observations in your dataset and specify
*fixed*versus*random*covariates in your model using*Linear**Mixed**Models*(LMMs). - How to generalize the distribution of the response variable from a normal Gaussian distribution to Poisson, Negative binomial, etc. using
*Generalized LMM*(GLMMs). - How to identify and define non-linearity among response and explanatory variables using
*Generalized Additive Mixed Models*(GAMMs).
## Learning Objectives- Use the nlme, lme, and mgcv packages to fit linear mixed models, generalized mixed models, and non-linear mixed models.
- Add a random structure to nest observations and identify non-independence.
- Use the
*family*function to change the distribution of the response variable when assumptions of linear regression are violated. - Add a smoothing term for the explanatory variables to induce non-linearity.
- Understand model outputs in R.
- Compare models and identify the most parsimonious model.
- Plot models in R.
## Prerequisites
If you would like to bring your own laptop to work on, please install R and R-Studio on your own before the workshop to speed things up. If you are having difficulty with this step, come to the workshop at least 20 minutes before it starts and we will help you out. |

Workshops >