Paul Schrimpf

For teaching an IO course, I [wrote a simple version of BLP using Turing.](https://ubcecon567.github.io/Bayes101.jl/quasi/) It doesn't include any supply side moments. For the same course, I made [a package for...

> @schrimpf These are pretty amazing! Would you mind if we adapted the BLP with Turing example you put up? Yeah, that'd be great.

What do you have in mind? `coef` inherits its element type from `y` and `X`. [These are constructed from the dataframe and formula, and converted to `Float64`](https://github.com/FixedEffects/FixedEffectModels.jl/blob/2a2661e508c579b6c9c8a82f1388debe0c8229e8/src/fit.jl#L173-L179). I could imagine...

I believe the segfaults in the tests from VectorizedReductions.jl and NaNStatistics.jl are caused by reducing over empty collections which leads to code along the lines of ```julia A = Float64[]...

I've been having similar problems with https://github.com/UBCECON567/juliabinder I suspect the issue is https://github.com/binder-examples/minimal-dockerfile/issues/8 . Based on https://github.com/matthewfeickert/failing-repo2docker-with-dockerfile, a possible fix is to use restrict jupyter-server < 2.0. I'm trying this...

I have written lecture notes in LaTeX, Rmd, RST, and Weave.jl. I would say that Rmd is my favoriate system. For me, the single most important feature is Rmd's [chunk-caching](https://yihui.org/knitr/demo/cache/)....

I believe issue here is that when reducing over an empty collection, code like the following is generated: ```julia A = Float64[] out = zero(eltype(A)) @turbo for i in eachindex(A)...