Andreas Noack

This would be great to have.

Any idea why CI still installs `[email protected]` even though it was yanked four days ago https://github.com/JuliaRegistries/General/pull/82386

I think we might need to add a branch for this case. While the quadratic form will always go to infinity as one of more elements diverge when the matrix...

In the method I linked to, I think we'd have to check if the vector argument in the quadratic form for any non-finite elements. I think we can assume that...

Indeed, I started playing around with PRIMA. It looks like the PRIMA COBYLA works here while PRIMA's BOBYQA seems to have inherited the issue from Powell/nlopt, see https://github.com/JuliaStats/MixedModels.jl/issues/705#issuecomment-2980446198. I should...

Indeed. I've tested with the new version and this issue seems to be resolved.

Indeed. Would you be able to prepare a PR?

Given that, according to the docstring https://github.com/JuliaStats/Distributions.jl/blob/a1010e408dbbdd4bcfd9c11eef189df64f7bb05a/src/common.jl#L96-L102 the sole purpose of `eltype` is to return the type of `rand`, which I agree isn't really feasible, should we also deprecate the...

Looks like people have hit the same issue over at StatsPlots, https://github.com/JuliaPlots/StatsPlots.jl/blob/e34f65ee293fc35678a3abc9d45fd74a65918c22/src/boxplot.jl#L79. However, this whole modificiation business that has been inherited from the original implementation seems dubious to me. I'll...

Thought I'd better get it straight from the horse's mouth so here is Tukey: i.e. he suggests that the whiskers end at the closest outlier. I think that makes sense....