Phillip Alday
Phillip Alday
Will #545 help with this?
For tinkering and comparison to lme4, look at this [gist](https://gist.github.com/palday/7b6f3fa43a1d5df4dbc2051e682d52d3).
After pulling my hair out for far too long trying to figure out why fast fits worked but 'slow' ones didn't, I tried a different optimizer. It turns out that...
Based on tinkering elsewhere (e.g. toying with fitting a Gamma model with identity and log links to the `languageR::lexdec` data -- there was a proposal to use Gamma with identity...
For tests, yes, because there could in theory be further optimizations to the MersenneTwister (or more specifically, methods that interpret the stream produced by MT). On the PRNG front: I'm...
I'm open to this, but: 1. I'm wondering if we should keep `GLMresp` around for some surface compatibility with `GLM.jl`. Of course, I don't think we really have any other...
There is a theoretical difficulty which makes it hard to implement in MixedModels.jl. lme4 can handle the gamma family, but note a few things: 1. There are numerical and theoretical...
@kirimaru-jp The fact that it's GPL'd and so any derivative work must be GPL'd, but MixedModels.jl is MIT-licensed, may pose a problem. It looks like @stevencarlislewalker is the only contributor...
We hadn't given much more documentation previously because we had only encountered it in a few places in the wild and `Grouping()` works like any other contrasts / hints for...
That would be great.