Probabilities Of Causation: Three Counterfactual Interpretations And Their Identification

By Judea Pearl. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Improving the accuracy of medical diagnosis with causal machine learning (2020, neutral): “Note the use of suf ﬁ ciency here is in line with 60 and does not refer to the INUS conditions \cite{mackie1974ce-ment}.”
Probabilities of causation: Bounds and identification (2000, methodology): “…the three probabilities of causation are bounded as follows: max[0; P (yjx) P (yjx0)] PNS min[P (yjx); P (y0jx0)] (20) max[0; P (yjx) P (yjx0)] P (yjx) PN min[P (yjx); P (y0jx0)] P (yjx) (21) max[0; P (yjx) P (yjx0)] P (y0jx0) PS min[P (yjx); P (y0jx0)] P (y0jx0) (22) [Pearl, 1999] derived Eqs.”
Causal superseding. (2015, background): “Existing work on counterfactual theories of causation suggests that people regard an event as a cause of the outcome when it satisfies two counterfactual conditions, 'necessity' and 'sufficiency' (e.g., Pearl, 1999; Woodward, 2006).”
Debugging Machine Learning Tasks (2016, methodology): “In contrast, Psi is based on a formal notion of causality introduced by [8, 7], and uses this eﬀectively to isolate training data slices that are responsible for test errors.”
Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Models (2019, methodology): “In particular, the monotonicity assumption (Pearl, 2000; Tian & Pearl, 2000) is sufficient to identify the Probability of Necessity and related quantities used in epidemiology (Cuellar & Kennedy, 2018; Yamada & Kuroki, 2017) to answer counterfactual queries with respect to a binary treatment and out”
From counterfactual simulation to causal judgment (2014, methodology): “Our counterfactual simulation model shares with Pearl (1999) the idea that people compute the relevant counterfactuals by manipulating an intuitive causal domain theory.”