Reproduce “Constraint-Based Causal Discovery using Partial Ancestral Graphs in the presence of Cycles” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Joris M. Mooij, Tom Claassen
The authors show that Fast Causal Inference (FCI) can remain valid when systems contain feedback. They prove that acyclic representations preserve the independences FCI uses, producing an ancestral graph for an equivalence class. With simple causal models, σ-faithfulness, no selection bias, and consistent tests on i.i.d. data, FCI is consistent, but does not generally identify a unique graph or establish cycles.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.