Vincent Arel-Bundock

@strengejacke I believe that your commit https://github.com/easystats/insight/commit/87c092758ebaba97b8fbd029afe54b4418a8a8d8 broke the Satterthwaite DF estimation for `get_predicted()`. When we call `get_df()`, we now get one DF per coefficient. This is not what the...

OK, but how do I get the per observation degrees of freedom for use in my own functions in marginaleffects? Those used to be the default. Are they still exposed...

> Ah, got it. I thought you were using `get_predictions()` directly. I can add an argument that will return the per observation df's. But I think the default should be...

I'm on the move, so I can't test it now, but at firdt glance this looks great. Thanks! The last marginaleffects was released just a few days after the last...

Looks like it's working now. Thanks again, I really appreciate the help on this!

Yes, everything should be fine. I'm no longer running tests on CRAN...

I agree we should improve things in that area. A few thoughts: 1. I don't think we should automatically go from `HC` to `CR` as those are different things, statistically...

Right. I just thought the spirit was to deprecate `robust`, because it is a confusing labelled used in a ton of different ways. Instead, the `vcov` refactor was meant to...

My initial plan was to do the same thing we do for `MASS::polr`: ``` r library(insight) library(MASS) polr(factor(gear) ~ mpg, data = mtcars) |> get_predicted() |> as.data.frame() |> head() ```...

The original issue has been fixed by commit https://github.com/easystats/insight/commit/fac472d0ce7cba826ca926e77dec908e38bde5ae I changed the title of the issue to reflect a separate feature request made [in a series a post starting with...