Multinomial Logistic Regression Packages | Tips

Good Day Julians:

I have looked several places including ATI (Alan Turing Institute)
to find a package since 2019 that help me perform a multinomial
logistic regression on the following DF:

Scale = ["A", "B", "C", "E"]
DF = Dataframe(ID = 1:1:15, M1 = rand(0:5:100,15), M2 = rand(0:5:100,15),
                          M3 = rand(0:5:100,15), Out1 = rand(Scale,15), Out2 = rand(Scale,15))

What packages exist that could help me identify
associations between the exploratory variables
[“M1”, “M2”, “M3”] and the response variables
[“Out1”, “Out2”]?

Also, pre-empting performance bottlenecks,
what are the dimensional limits for any module
you present? For example, if I had a DF sized
(10,10) with two categorical response variables
and float/int typed exploratory variables, might
I have convergence issues after a certain
number of iterations or dimensional size?

Thank you,

Econometrics.jl has multinomial logistic regression I believe.

2 Likes