That’s not correct! 👇🏻 The assumption is much milder. We just need to assume that the factors we condition on (or stratify on, in epi lingo) block all chains of dependency that could connect the exposure and the outcome. A minimally sufficient adjustment set may not include unobserved confounders. @yudapearl

Adan Z Becerra, PhD @AdanBecerraPhD

2 months ago

That’s not correct! 👇🏻 The assumption is much milder. We just need to assume that the factors we condition on (or stratify on, in epi lingo) block all chains of dependency that could connect the exposure and the outcome. A minimally sufficient adjustment set may not include unobserved confounders. @yudapearl

1 0 0 2K 0

@soboleffspaces Sure I was a bit sloppy. Indeed we don’t need to adjust for every confounder but we do need to adjust for every confounded chain.

@soboleffspaces Agree. A minimally sufficient adjustment set may not include any confounders.