Phillip Alday
Phillip Alday
Yes. And I think it makes sense to have the StatsBase API just redirect to various properties. That way people can still abstract away via the functional interface when e.g....
I think the question should be: what do we want for the default? Because we can also support additional methods `dof_residual(::LinearMixedModel; variant::Symbol)` where `variant in (:leverage, :ferank, :nparam)`. At one...
Looking at the remaining items, here are my thoughts: - Cook's Distance: if I recall correctly, this can be expressed in terms of the diagonal of the hat matrix. Will...
As @dmbates indicated in his Discourse comment, this isn't a trivial fix, but I started [working on it](https://github.com/JuliaStats/MixedModels.jl/tree/pa/multimember) a few weeks ago. I haven't figured out a good formula syntax...
Interesting. When we get to real benchmarking, I would be very curious whether how GLMM does with the gradient information.
That unfortunately is what I was afraid of, once I actually started to think about it.
Mostly a note to myself here, but we still have the indirect dependence on `Distributions` via `GLM`. There is also the Chisq CDF in `StatsFuns` as an alternative for computing...
Internally a [new model is constructed to simplify the bookkeeping][1] and model construction relies on the general machinery for constructing model matrices, which then requires being able to compute the...
@jariji With the help from @kleinschmidt I've found a simple fix for this error, which we should be releasing shortly (see #653).