- Concentrate more on nested random effects [x]
- Drop or move t-distribution version and ordinal version to later [x]
- Drop GPs and put in change-points instead, and also perhaps time series
- If you're putting in an example make sure it runs and you have the plots [x]
- Do a better job of explaining the zi model in Stan. Really complex [x]
- For varsel, add in plot of un-regularised vs regularised
- Add in better data set for looking at splines
- Change occurrences of
dbin(p, 1)todbern