nonlinear parametric modeling post
Closed this issue · 0 comments
LucyMcGowan commented
I think it would be cool for our next post to be on modeling nonlinearly (using restricted cubic splines). A couple of points
- Let's bring Gaussian back - the common nomenclature for models can be confusing -- linear regression having non-linear effects. Why don't we call that Gaussian regression?
- Grambsch and O’Brien show that testing for linearity is a degrees of freedom disaster!
- Fitting flexible nonlinear parametric models reduce the need for many cases of model checking
Another note: it would be cool if we could find a way to add "notes" to the posts, so like if I wrote a post you could add some "notes" throughout the text that would clearly be from you (maybe right set and a different color?) and vice versa.