arrowhead_precision: Arrowhead Precision and Recall

In tzimiskes/causality: Examine The Causal Structure of Data

Description

arrowhead_precision calculates the arrowhead precision between two causality graphs

arrowhead_recall calculates the arrowhead recall between two causality graphs

Usage

arrowhead_precision(x, y)

arrowhead_recall(x, y)

Arguments

x	A causality PDAG
y	A causality PDAG

Details

arrowhead_precision counts the number of directed edges ("-->") in x and then counts how many directed edges in x are also in y. Then, the ratio is returned.

arrowhead_recall counts the number of directed edges x and then counts how many directed edges in y are in x. Then, the ratio is returned. 1 implies that every directed edge in x are also in y.

Value

Length one numeric between 0 and 1. arrowhead_precision returns NA if there are no oriented edges in y.

arrowhead_recall returns NA if there are no oriented edges (arrows) in x

Author(s)

Alexander Rix

References

Joseph D. Ramsey: “Scaling up Greedy Causal Search for Continuous Variables”, 2015; arxiv:1507.07749[cs.AI].

Spirtes et al. “Causation, Prediction, and Search.”, Mit Press, 2001, p. 109.

See Also

Other graph comparison statistics: adjacency_precision, adjacency_recall, and shd

Examples

TODO(arix)

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