Yesssss! Thanks so much for giving this a shot. I hope you like it! And please feel free to let me know if you hit a roadblock or find an error.

OK, this is amazing. I saw the first one in roughly 20 seconds. One of the skills I was most proud of back in the day!

ok, i laughed

Working on an R&R now. A few weeks away from submission. Will let J decide if we update the OSF version between re-submit and final decision.

svg

Aww thanks! Wait for the upcoming revision. Julia has done a ton of work on it, and I feel the next draft is going to be much better.

Aaaww, so glad to hear that. The best compliment I could hope for.

Nice! Makes me think of the crater they flooded in the north of Quebec to build the Manic 5 hydro-electric dam. maps.app.goo.gl/5oiCmb3avy8X...

This looks great. Congrats and good luck!

Not sure. I looked at those a long time ago. Let me know if you give it a shot and have questions. One thing to note is that `marginaleffects::comparisons()` builds very simple contrasts, where one or a few variables change uniformly. Time series "shocks" or matrix predictors can be more complicated

Wow, that's too kind! And thanks for ordering. You too can now have a pile of books like this one.

Categorical responses are not inherently problematic. It's a tiny bit more challenging, but not much. The real constraint is that the modelling package must supply all the standard coef(), vcov(), and predict() methods. See here for details: marginaleffects.com/bonus/extens...

Glad you think it's clarifying!

Great tables is the {gt} port. FWIW, I think it is excellent, and that's what I use in Python. At some point I started writing code for a {tinytable} port, but it was too much work and I gave up.

Nice! I was just in Ottawa and visited her giant "Maman" spider.

love it!

[Her apple crumble is also the best]