Generalized Additive Models
In this week we’ll look at breaking the assumption that the covariates are linearly related to the inverse link of E(Y).
- Detect violations of the linear assumption in residual plots
- Use a penalized smooth spline term to fit an arbitrary non-linear function
- Decide what type of model to use based on the properties of the data
Here’s a handout that I call putting it all together. I’m not sure when the best time is give you this handout, but now seems to be a good time. In the first part, which deals with deciding what sort of models to fit, when you get to step 6 just answer NO and you’ll be good. Otherwise you will probably understand what the answer yes means in a few weeks. The 2nd part is my guidelines for model selection. These are my opinion and a fair facsimile of my usual practice; they should not be taken as absolute rules supported by independent statistical theory.
Optional online help session 10-11, read, watch videos, homework, discuss the handout if needed.
Before Wednesday’s class