bachlaw

Would appreciate it if categorical and/or multinomial CDFs could be added to the math library. It looks like folks have had success using truncated Poisson variables to generate at least...

I note this suggestion on the Discourse of a multinomial probit implementation that might be worth pursuing: https://discourse.mc-stan.org/t/multinomial-probit-in-stan/17583

Adding `-1` to the model formula has helped to solve this problem for me, suggesting that something about the Intercept or the perceived need for a column reflecting one in...

This would be a very welcome addition. `glmTMBB` has a natural implementation of many of these structures which seem to work (exclusively?) at the group level, and anything from the...

As a Bayesian example would probably be more helpful, [INLA ](https://inla.r-inla-download.org/r-inla.org/doc/latent/ar.pdf) accommodates this through their notation: `y ~ 1 + f(time, replicate=group_idx, model='ar', order = 1)`. However, it is possible...

Paul, I agree this would be a terrific add. With `glmmTMB`, the generalized poisson provides very good fits fairly quickly to overdispersed data, and as you point out can theoretically...

McCulloch & Neuhaus wrote a very good article a few years ago claiming that "mis-specifying" random effects models with a normal distribution was essentially a non-issue, and contending that arguments...

I appreciate having the option to specify a different distribution for the random effects. In some of my datasets, however, it would also be useful to accommodate meaningful outliers with...

> At the moment it's only hard-coded for gaussian, poisson, bernoulli, and binomial marginals, but is compatible with any distribution that has a CDF defined - so these marginal functions...

Indeed: I noticed that in Stan there generally don't seem to be CDFs for discrete distributions other than binary outcomes. Perhaps there is a comparable implementation in `scipy` or a...