Jacob B. Socolar

For some of my models with a lot of parameters, treedepth has almost no variability, even if the step-size is right near the border between two treedepths. Check this out:...

I don't think that `ar()` plays nicely with Bernoulli outcomes, as the variance isn't well identified?

Let me know if this actually needs to be filed against math or somewhere else.

Just to chime in that by the time you get up to 1e+5 parameters, the `readLines` approach takes hours at least, while the `fread` approach remains fast. Tagging @bgoodri and...

Hi @paul-buerkner @SimonCMills just rewrote `rstan::read_stan_csv` to use `data.table::fread` and achieved speed-up on large inputs in line with what we expected. That's available here: https://github.com/SimonCMills/rstan/blob/develop/rstan/rstan/R/stan_csv.R I've been working with Simon...

We're waiting to hear back from `rstan` devs only because pulling out all of the `rstan` internals behind `rstan::read_stan_csv` would be a bit of a pain. If `rstan` is slow...

Just want to note that spatially varying coefficient models are already possible in `brms` via nonlinear formulae. Here's an example using a Gaussian Process (because that's more intuitive for me...

Sorry, I might have spoken too soon about the extensibility to autoregressive structures. For me, the following ```r # make a symmetric adjacency matrix M

Just want to flag that 2.32 release date is tentatively planned for 17 April, so this is becoming more urgent. See https://discourse.mc-stan.org/t/planning-the-2-32-release/30641 Edit to add: the urgency is just that...

No action on the rstan side, but a function that achieves exactly this has been implemented in `brms` here https://github.com/paul-buerkner/brms/pull/1400