Aki Vehtari
Aki Vehtari
> Is line looR2[looR2 < -1] no CV (i.e., test dataset = training dataset) We have used Bayesian-R2 for that as it has some benefits in that case, but the...
Implemented by https://github.com/stan-dev/projpred/pull/496/commits/0ed83914eff49bba52c6c065e7059d0b300b9643
> For the negative binomial distribution, I haven't been able so far to simplify this expression to a more tractable numerical solution would be fine, too > 2. It is...
I have McCullagh and Nelder (1989) and the definition is what I remembered. I'm now confused what is the problem in Piironen et al (2020).
Thanks for creating the issue
Improved plot examples: - intervals  - km_overlay  - PIT-ECDF  The colors are not specifically the best, but otherwise I like these
When I made the issue, 1) PSIS resampling was often failing in CmdStan, 2) there was no option for sample without replacement which would be better for getting unique initial...
Birthdays case study illustrates the use of the above functions https://users.aalto.fi/~ave/casestudies/Birthdays/birthdays.html
Closed by #937
I was also annoyed the Pathfinder expects a list and not list of lists, but I forgot to make an issue. I agree, it should be possible to give a...