Causal inference and causal explanation with background knowledge

By Christopher Meek. 1995

This paper presents correct algorithms for answering the following two questions; (i) Does there exist a causal explanation consistent with a set of background knowledge which explains all of the observed independence facts in a sample? (ii) Given that there is such a causal explanation what are the causal relationships common to every such causal explanation?

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

Identifiability of Causal Graphs using Functional Models (2011, background): “[2001], Meek [1995] show how one can exploit conditional independences in the data for (partially) reconstructing the graph Gc, e.”
Order-independent constraint-based causal structure learning (2012, methodology): “Any ordering leads to the removal of both edges, as the separating set { X 1 , X 3 } for ( X 4 , X 2 ) is contained in a ( X 4 ) , and the separating set { X 1 , X 5 } for ( X 3 , X 4 ) is contained in a ( X 3 ) (and in a ( X 4 ) ).”
Structural Intervention Distance (SID) for Evaluating Causal Graphs (2013, background): “We extend each chordal chain component c (see Section A) locally to all possible DAGs Cc,1, . . . , Cc,k, leaving the other chain components undirected (Meek, 1995).”
A characterization of Markov equivalence classes for acyclic digraphs (1997, background): “The essential graph associated with an (equivalence class of) ADG(s) was first introduced by Verma and Pearl (1990) as the completed pattern associated with the ADG.”
Masked Gradient-Based Causal Structure Learning (2019, methodology): “Constraint-based methods, such as PC and fast causal inference [33], use conditional independence tests to ﬁnd initial skeleton and then determine the edge directions according to certain orientation rules [19, 38].”
Nonlinear directed acyclic structure learning with weakly additive noise models (2009, methodology): “∈ Ψ, directing all Vj ∈ Pa Vi G as Vj → Vi and all Vk ∈ Ch Vi G as Vi → Vk, and (iii) applying the extended Meek rules [4], treating orientions made using Ψ as background knowledge.”
Detection of Unfaithfulness and Robust Causal Inference (2008, background): “These rules are shown to be both sound and complete in Meek (1995a).”
Bayesian Networks for Variable Groups (2015, methodology): “After v-structures have been directed, one can direct the rest of the edges that point to the same direction in every DAG of the Markov equivalence class using four local rules often referred to as the Meek rules (Meek, 1995).”
Causal inference in partially linear structural equation models (2016, methodology): “Orientation rules R1–R4 for Markov equivalence classes with background knowledge from [12].”
A Scoring Function for Learning Bayesian Networks based on Mutual Information and Conditional Independence Tests (2006, methodology): “…and search, although there are also algorithms that use a combination of independence-based and scoring-based methods with different hybridization strategies (Acid and de Campos, 2000, 2001; Dash and Druzdzel, 1999; de Campos et al., 2003; Singh and Valtorta, 1995; Spirtes and Meek, 1995).”
Structural Intervention Distance for Evaluating Causal Graphs (2015, methodology): “We extend each chordal chain component c locally to all possible DAGs C c , 1 , . . . , C c , k , leaving the other chain components undirected (Meek, 1995).”
Graphical criteria for efficient total effect estimation via adjustment in causal linear models (2019, background): “…f r o m h tt p s :/ / a c ade m i c . oup . c o m / j r sss b / a r t i c l e / 84 / 2 / 579 / 7056108 b y R o y a l L i b r a r y , C ope nhagen U n i v e r s i t y L i b r a r y u s e r on 13 A p r il 2023 Bühlmann, 2012; Hoyer et al., 2008; Meek, 1995; Scheines et al., 1998; Wang et al., 2017).”
Identiﬁability of Causal Graphs using Functional Models (methodology): “Given that one observes i.i.d. data sampled from ( X i ) ,i ∈ V , Pearl [2009], Spirtes et al. [2001], Meek [1995] show how one can exploit conditional inde-pendences in the data for (partially) reconstructing the graph G c , e.g., by using the PC algorithm.”