Alex Arslan

Unfortunately with fused broadcast there doesn't appear to be an easy way to specialize on a combination of functions, so we'll have to find another way to make `.!isna.(x)` return...

I don't think we should remove the implicitly vectorized `isna` deprecation, since then we're in conflict with how things are done in Base. Until we can figure _something_ out, be...

Yes, for sure, adding the log-rank test to this package has been on my backlog of desired features since about 2017. 😄 Apologies, I should have spoken up on this...

Hey Pietro, that sounds like an awesome idea. If you'd be willing to contribute that functionality to this package, that would be awesome!

`EventTimeVector` is indeed intended to be a performance improvement over `Vector{EventTime}`, since it stores the survival times as a contiguous array of scalars and censoring indicators as a `BitArray`, rather...

Ah right, formulas are an important part of things like this... That functionality will eventually be split out of DataFrames but in the meantime that's where it lives so I...

> On a separate note, when the formula formalism gets an independent package, it will be important to document it I believe StatsModels has some docs, and I thought there...

Can we just let this package die with 1.0 with a clear note in the README?

For Cox models, you mean?

Excellent list! I'm not sure when I'll have time to do the refactoring to avoid duplication between Kaplan-Meier, cumulative incidence, and Nelson-Aalen, so a PR for that would be welcome...