chickering: Turn a causality DAG into a pattern.
In tzimiskes/causality: Examine The Causal Structure of Data

Description Usage Arguments Details Value References

chickering converts dags into their markov equivalence class. This function exists mostly to provide documentation on how causality dags are converted to patterns (or pdags). Using as.pattern instead is recommended.

1	chickering(graph)

graph

a causality dag.

chickering generates a topological sort of the dag, and generates an ordering of the edges based off this topological sort. Then, every edge is marked as 'unkown' and algorithm iterates through the nodes acoording to their order. If there is an unshielded collider on a node, all of its parents are marked as directed.

a causality.pattern

Chickering DM. A transformational characterization of equivalent Bayesian network structures. InProceedings of the Eleventh conference on Uncertainty in artificial intelligence 1995 Aug 18 (pp. 87-98). Morgan Kaufmann Publishers Inc. arXiv:1302.4938 [cs.AI].

tzimiskes/causality documentation built on Sept. 15, 2019, 8:41 p.m.