Seth Axen

> Thanks @sethaxen. The LKJL here is in the very early stages, so I think this is very likely an issue with the implementation itself, so I've moved the issue...

Actually, I think there is an error in the `logdensity` from `LKJL`, which was carried over from AltDistributions. Compare with https://github.com/JuliaStats/Distributions.jl/issues/1336 (I've verified the Jacobian determinant there with FiniteDifferences) and...

There's also using expectations of basis functions. E.g. characteristic functions. Useful for defining a measure as a convolution of measures.

Also one might want to define a measure from some function that computes random samples.

> This one will probably be tricky, maybe @zenna's work could help with it? Yeah, I'm just listing things a user might want to do, not necessarily that will be...

Many of these I don't know much about, but I'm very excited about the ones I do. Where would changing the measure fall in this list? e.g. One has a...

We should probably use whatever the standard notation is for product measures (I don't know what that is). I also feel strongly that unicode symbols as operators should only be...

TBH I haven't thought about this at all from a design perspective. For the moment I just need a way to represent a uniformly weighted mixture of multivariate normals. I...

The purpose of this PR was to provide an easy way to normalize a measure, but I don't think this is the right approach. For example, suppose we have a...

@cscherrer if you're thinking of making a BayesianOnlineStats package I'd be happy to contribute. It'll be a good excuse to spend more time thinking about how to work with streaming...