Andreas Noack

In my opinion, these should all be in `Distributions`. We need `Rmath` to cover cases where we don't have good samplers yet but hopefully we can just switch the `rand`...

@ron-wolf Bump. Would you be able to add a tests as suggested in https://github.com/JuliaStats/Distributions.jl/pull/1159#issuecomment-702589660?

I've updated the list

Just a minor comment here. So actually the issue with `eltype` for DataArrays has been fixed in version 0.7.0 but unfortunately, it will probably be difficult to ever install that...

@YongHee-Kim You might be able to see why `DataArrays` isn't updated by executing `Pkg.update("DataArrays")`.

Currently in `IndexedTables`, categorical-like data is represented by `PooledArrays`. They are simpler than `CategoricalArrays` mainly because the elements aren't special `CategoricalValues`. You might want to try and see if `PooledArrays`...

Not that I'm aware of.

> Isn't the case where both a and b are zero undefined? I think so but I might have formulated it poorly. It wasn't that case I was interested in....

Possibly related to this. If I use a numerical grouping variable in the examples in https://github.com/JuliaPlots/AlgebraOfGraphics.jl/issues/372 then I get ```julia julia> draw(data(df) * mapping(:x, col=:gr) * histogram(), facet=(;linkxaxes=:minimal)) ERROR: MethodError:...

This seems like a great improvement to me (and it fixes a clear bug). I'm not familiar with this code so I'll let others review.