Presents LocPC and LocPC-CDE algorithms that learn a local essential graph using only local conditional independence tests to identify controlled direct effects under weaker assumptions than global PC methods.
This guarantees thatCDEpx,x 1,yqisnotidentifiable, according to Theorem 5.4 of [Flanagan, 2020]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Local Markov Equivalence for PC-style Local Causal Discovery and Identification of Controlled Direct Effects
Presents LocPC and LocPC-CDE algorithms that learn a local essential graph using only local conditional independence tests to identify controlled direct effects under weaker assumptions than global PC methods.