Douglas Bates

@palday BLAS wasn't quite the issue that I needed to work around recently. What was happening to me recently is that compiling the lme4 package for R caused it to...

Can you expand on what you would like such an API to be?

It appears that this will need several methods added to the `LinearAlgebra` package to do things efficiently with `mul!`. Let's defer this until v3.0 is out.

Not really a schedule, per se. I haven't written the code yet and I need to be very careful about proper tests, etc., when making a PR on a standard...

It may be better to return a `Vector{DictTable}` rather than a `NamedTuple{DictTable}` because the expression for the grouping factor is the name of the keys in the `DictTable`. Update: not...

Thank you for opening what is indeed an interesting issue. I can see that the results from [`lme4`](https://github.com/lme4/lme4) and from this package do not coincide. What I don't know is...

Thanks for the quick response, Ben. I have a vague recollection of discussion related to the dispersion parameter for the `Gamma` family and how its value influences the evaluation of...

@hikea You write "Regarding raw_rt, I did convert lexdec RTs back to milliseconds, so all my examples are on raw RTs." I'm not sure what that means. Can you be...

@hikea Thanks for the explanation. I went back and looked at the documentation for `languageR::lexdec` and found that the `RT` column is logarithmic. Using `lexdec[:rt_raw] = exp.(lexdec[:RT])` I get a...

Note that the deviance as reported here with the inverse link is slightly better (i.e. smaller) than the corresponding fits with the identity link. I say `as reported here` because...