jtextor
commented
4 years ago This is intentional. There doesn't need to be any relationship between X and Y for there to be a confounder -- in a graphical sense, adding an X->Y arrow to the DAG does not affect the set of confounders at all and this is one of the things this tutorial is meant to emphasize
Love this package, and the web interface! Thank you so much. I’ve shared it with colleagues.
While clicking around, I was presented with this challenge.
It is solvable as far as the question being asked goes. However, it seemed unintuitive to me that X was not depicted an ancestor of Y. Perhaps that relationship should be enforced when a graph is generated? Or, perhaps this is intentional in some way?