I think Discord can be great! But will be inherently different, one of the parts that is still not quite anywhere else yet is all of the fancy people were in the same space (and not just for altruistic reasons!) as the newbies, so it was an amazing place to connect across levels

😢 I miss it too! I have so much to thank #rstats twitter for! Definitely my postdoc with Jeff, internship with Jenny& getting to hang with you at unconfs! I was just telling Noam how much I miss those. I feel like one problem now is there doesn’t feel like *one* place I know everyone will be talking

If I understand the use case I think ggdag can do this too! r-causal.github.io/ggdag/refere...

🙏

Some corrected headlines 🙏

🙈

(So in my case the Xs tend to be baseline characteristics measured at time 0, y would be something measured at some time after 0)

I think in my research it is very unusual for someone’s Y to be my X if we otherwise have the same Xs because I’m thinking of estimating causal effects where temporarily Y must follow X but maybe there are lots of use cases I’m not thinking about!

This is interesting I think I would have come to the opposite conclusion that more often in practice we have our Xs not Y so deterministic may be more practical of the two!

I will!!

I know 🥲 she’s to talented!!

Wake Forest also has a medical school & recently joined with Advocate Health making us the nation’s third largest nonprofit integrated health system — a huge opportunity if you’re interested in big (Huge!) data/ EHR /pragmatic trials etc. Several of us have joint appointments in the medical school 🤝

We have 9-month contracts (Aug - May) & are on “hard money” in a Statistical Sciences department — many of us cover the summer on external grants (although this isn’t required!)