Chad Scherrer

Consider this model: ```julia julia> m = @model begin N ~ Poisson(10) x ~ Normal() |> iid(N) y ~ For(N) do j Normal(x,1) end end; julia> predictive(m, :x) @model x...

Make `iid` take a shape instead of length

1

@vasishth asked on Twitter about implementation of models like [these](http://www.ling.uni-potsdam.de/~vasishth/pdfs/SorensenHohensteinVasishth2016.pdf). The "varying intercepts, varying slopes" model in Listing 6 would be something like this: ```julia # varying intercepts, varying slopes...

I accidentally included a recursion in a model today: ```julia m = @model begin μ ~ Uniform(-2,3) x ~ Normal(μ,x) end nuts(m(), (x=0.2,)) |> particles ``` And it seemed to...

Based on Tamas Papp's point [here](https://discourse.julialang.org/t/parallel-mersenne-twister/27567/12?u=cscherrer). Some possibilities: - Have a global default, and allow users to select others by calling a different method - Pass PRNG as an optional...

TODO: edit this with more details - just a quick note to myself for now... For a while, the `For` type was parameterized. At a point, this broke things and...

Translate Stan DiffEq code

2

Code is [here](https://github.com/JuliaDiffEq/DiffEqBayes.jl/blob/master/src/stan_inference.jl#L51-L83)

Differential Equations

8

Implement SDE, etc using [DifferentialEquations.jl](https://github.com/JuliaDiffEq/DifferentialEquations.jl) and (most likely) Stan-like approach

Feed the return values of one model into the parameters of another

Thanks for making CompatHelper :) There's a minor issue I see coming up a lot for my workflow. After CompatHelper creates a PR, I'll independently make some change that makes...