Write a FAQ man page

[nicholasjclark]

The {mgcv} equivalent is very useful, though not widely known about: https://www.rdocumentation.org/packages/mgcv/versions/1.9-1/topics/mgcv.FAQ. Things to cover

1. How to compare models using loo_compare() and lfo_cv() (loo less reliable for models with autoregressive processes; see also here; use in initial model development and comparison before adding latent dynamics, or stick with lfo)
2. How to extract posterior estimates
3. How to model seasonality (options include cyclic smooths and discussion of knot placement, or fourier terms)
4. How to model time-varying seasonality, and time-varying periodicity
5. Dealing with irregularly sampled time series (options include GPs and continuous time AR models)
6. How to report results from mvgam() models (point to the blog post on using {marginaleffects} but caution that this doesn't yet work with linear functionals)
7. How to check model assumptions (residual plots and ppc_check())
8. How to choose k