I’d like to start my Julia contributions to the StatsFuns.jl or the Statsistics.jl packages, specifically, transformations. Is it possible to add a boxcox or yeojohnson transformation to one of the packages? If so, what would be the best way to do so.
Doesn’t fit in Statistics.jl, which covers the basics and there’s already BoxCoxTrans.jl
Would it make sense for transformations to be included under one package? Why do we have a StatsBase.jl, StatsFuns.jl, and Statistics.jl? Then a typical transformation like box cox and yeo Johnson be in separate packages? I asked these questions for education purposes so I can understand how the Julia community have done these things.
I’m afraid there is no more principled answer than historical accidents and path dependency. Statistics is a standard lib and as such its development was tied to Julia versions, while StatsBase evolved outside of base. There were many discussions about somehow unifying things over the years that came to nothing.
Given that Statistics is now uncoupled from base I’d start there - open an issue and see what the maintainers think.