Feb 5-6, 2018
At this point we know how to specify, execute, and interpret output from linear models. But how do we know if they’re even working?
This week we will focus on model specification - how to determine what covariates to include, and how to assess whether the model has fit your data well.
- Model validation
- Standardization of covariates
- Variance Inflation Factor
- Model selection
As always, this class will have a mixture of R code and lecture slides
Full comprehension is not expected at this stage.
- McGill, B. (2015). How many terms should you have in your model before it becomes statistical machismo?
If you are interested in model averaging (which we won’t cover in this course)
Information theoretic approach example
Fukuda Y, Manolis C, Saalfeld K, Zuur A (2015) Dead or Alive? Factors Affecting the Survival of Victims during Attacks by Saltwater Crocodiles (Crocodylus porosus) in Australia. PLoS ONE 10(5): e0126778. doi:10.1371/journal.pone.0126778
Slides available on Speakerdeck