Patrick Baylis

Related: the [sf](https://github.com/r-spatial/sf) package geometry columns (which are list-columns) that presently cause write_fst to fail with error message `Error in write_fst() : Unknown type found in column.` Happy to open...

Both random access (for sampling) and on-disk subsetting would be very useful features for my use cases. Too bad about serialization but so it goes. I'm not an expert in...

Following up, this seems to be some sort of corruption issue with some of my .FIT files. So fixing it in this package may be out of scope. Fortunately there...

Thanks, that (the order of magnitude improvement in models w/ FE explicitly included) is great news! You are right of course regarding needing the full VCOV to perform a proper...

I'm probably over-complicating them. Suppose you're estimating a quadratic model with unit fixed effects, like this:  A response function f(X) is just the predicted value of y across a...

I took a shot at cooking this up today, actually inspired by some [code](https://stackoverflow.com/a/35808071/5447959) @kendonB put together a couple years ago. Fun fact: that link goes to an SO post...

Yes, you've got it exactly! Brilliant. I think I don't understand what I'm talking about myself until I put it into code. Your insight about the baseline is clarifying for...

Not at all! This is super helpful feedback for me to put together a PR at some point. But it's not pressing and you have a lot going on right...

Ohh, yes - that does make sense. I've seen it done before, although I'm actually struggling to find an example right now. I suppose the question is if it's worth...

I like `relpred` too. Sold! On Mon, Mar 15, 2021 at 7:58 AM Laurent Bergé ***@***.***> wrote: > *On the extra baseline stuff.* > If you, an user of such...