pymc-labs/CausalPy

Proposal to add explicit information about causal identification to IPW docs

Opened this issue · 1 comments

Hi CausalPy team,

Thank you for the new update to cp 0.3.0!

It's great to see the package growing.

I was going through the new IPW docs page today and I was wondering if it would be worthwhile to add a short note in the intro section a short explicit note about when IPW methods can vs cannot lead to causal identification.

Perhaps that could be useful for less experienced users who try to understand if the method applies to their use case.

What are you thoughts?

Always happy to see a PR!

Just to note that we also have this section explaining ipw with respect to strong ignorability:
https://causalpy.readthedocs.io/en/latest/quasi_dags.html

I think either could be expanded as you suggest. Or linked explicitly...