Peter Ralph

Notes, discussing with @gregorgorjanc: 1. one could compute the covariance between *individual* traits, which would depend on the model of dosage compensation, or between *offspring* traits (e.g., as above in...

Note this from our paper: 

I've got this sorted over in #1623, by switching to "average over samples" rather than "sum over samples" and providing the correction to the diagonal for lack of self comparisons...

Well, thanks for the feedbac, @GabrieleNocchi, but could you be a bit more precise with your recommendations here? Let's see: 1. As linked to in [the README](https://github.com/petrelharp/local_pca), there are three...

Thanks. FWIW I do agree with you - more documentation would be useful, and documentation is certainly worthwhile. Unfortunately, I've got too much on my plate to write a whole...

I have had a go at making the docs for `corners()` more clear - see what you think, and feel free to suggest improvements? Click the "files changed" tab at...

Thanks for the comments! Let's see: - I don't want to change `i` to `k`, because `k` is the *total* number of columns; so `i` is standing in for an...

Oh, sorry - I also meant to say that your suggestion for a different format is a good one, but I don't want to change that at this point, as...

Ah, I like that idea. I guess I don't want to just add those lines without actually seeing what it looks like, and I don't have those data any more;...

> Since it's returning an iterator I don't think there's any point in trying to be clever I agree with this, *but* if you *did* want to be clever with...