Jonas Kristoffer Lindeløv

@Marin-Kitagawa Yes. Try it, I haven't found ways to enable it by default. You can enable it locally in Everything's in-app search but that doesn't transfer to Fluent Search.

Thanks, @jpzhangvincent, that would be great! I think getting it to work is simply a matter of (a) re-writing a few JAGS models as stan models and learn if they...

(y) I also need to glance over that list and write some more accurate titles. For example, most functions including `fitted()` and `predict()` can also work from prior distributions.

Negative Binomial regression is not natively supported yet, but I will make it a priority adding it. Specifically: - [ ] Support for `mcp(..., family = negbinomial())` with appropriate default...

@josswright Yes, mathematically and in JAGS it's straightforward to extend mcp to include multinomial models, as long as there is a continuous predictor. Just need mcp to count the number...

@jpzhangvincent I chose JAGS because: 1. Ease of installation. I have had many students struggle installing RTools. I myself recently tried installing RTools4 without success. 2. Total speed. Many problems...

@jpzhangvincent Yes, definitely! `fit = mcp(...)` returns an `mcpfit` object and you can see the JAGS code using `cat(fit$jags_code)`. This code is mostly generated in `R/get_jagscode.R` and `R/get_formula.R` using the...

Thanks for noting this! I'll update the doc. Do you know why `EnvCpt` and `cpt.meanvar()` do not find the same change points in the example?

Random change points now supported in a7dfbb4dfe2c44b728ce9be2f98e56dc52f41e83. Saving random slope and intercept for v0.2

@luc-w I'm currently working on consolidating the current features so it may be ½ year in the future.