Jeff Mills

Ok, no idea how to do that at present (trying to finish up a paper on predictive model selection at the moment). I think the few tests that are currently...

Hey Dominique! I just found all these comments! No, no good reason, just the first attempt. I have add bounds to keep it in [-1,1], but haven't thought about HDI...

Actually, for most of the tests implemented so far, since the distribution is unimodal and symmetric, the quantiles give the HPDI. For the correlation test, this will be true as...

To do asap: 1. Add ASH code for better kernel density evaluations. 2. Change function names to add bayes_ prefix. 3. Determine if separate HPDI function is needed yet (I...

Yes, excellent idea. I will go through functions and start doing this.

I like the bayes_ prefix. I think cortest or cor_test makes sense, not sure which looks better.

Yes, I agree. I have been meaning to look into using Documenter.jl to make a nicer set of docs, but haven't had time yet. I don't know how to make...

Yes, the plotting ecosystem seems to be a bit of mess since v1.0 also, so separating them out makes a lot of sense. Unfortunately I have almost no experience with...

Hey, I think this looks good, but I'm not quite sure what I do here. You have renamed a few functions, e.g. Bayes_correlation, which looks good, just not sure how...

Yes, I fully intend to do this sometime soon.