Douglas Bates
Douglas Bates
The traces also show me something about the first plot (the one that looks like a "bell curve" with upturned edges) that I didn't understand. That curve should not have...
We already have `StatsModels.jl` as a dependency so should we make this a method for `StatsModels.lrtest`?
I am considering whether it is best to try to explain why the methods or formulas from LMs or ANOVA that readers think should be meaningfully extended to LMMs aren't....
I agree
Fitting this model is a delicate optimization problem. Even slight differences in the floating point evaluations or order of operations in the numerical linear algebra could result in convergence to...
I am also on an arm macos device and I do have a `startup.jl` file that loads `AppleAccelerate.jl` which could cause a change. I will look into `Base.julia_cmd()`
> Malt is used to run the workers, and it does https://github.com/JuliaPluto/Malt.jl/blob/124c471c574d0958fd349858fd3686b798068ff3/src/Malt.jl#L255. Do we reckon that setting `"OPENBLAS_NUM_THREADS=1"` might have anything to do with this? Yes, that's the difference ```julia...
I think this issue can be closed as we know the cause of the discrepancy.
You can easily check the effect of multi-threaded custom BLAS with `peakflops`. It is nearly a 10-fold difference on an arm64 MacBook. ```julia % OPENBLAS_NUM_THREADS=1 julia --startup=no _ _ _...
> Wait, is the actual desired constraint simply a diagonal covariance structure? Something like the `||` in lme4? What is desired here is modification to the covariance structure of the...