c-components
A graphical structure concept introduced by Tian and Pearl within the causal inference framework, referring to sets of observed variables that are mutually connected via bidirected-edge paths in a DAG with hidden confounders. c-components are the core structural units used in the recursive derivations of the ID algorithm.
Timeline (last 90 days)
The main contribution of the paper is to identify a large class of clustering operations that satisfy identification invariance, and to provide a criterion centered on the c-components (confounded components) of the original graph.
C-components (confounded components) were introduced by Tian and Pearl in the causal inference framework, referring to components formed by observed variables that are mutually connected through bidirected edges (shared unobserved common causes).
The ID algorithm is a complete algorithm for causal effect identification, and each step of its recursive derivation decomposes the problem along the boundaries of c-components.
The more and smaller the c-components, the easier it is for identification algorithms to decompose step by step; if a large number of variables belong to the same c-component, identification often becomes difficult or even infeasible.
The core criterion of the paper is that a safe clustering operation must respect the topological structure of the c-components of the original graph and must not arbitrarily merge variables belonging to different c-components into the same cluster node.
All Facts (5)
The main contribution of the paper is to identify a large class of clustering operations that satisfy identification invariance, and to provide a criterion centered on the c-components (confounded components) of the original graph.
50%UnverifiedC-components (confounded components) were introduced by Tian and Pearl in the causal inference framework, referring to components formed by observed variables that are mutually connected through bidirected edges (shared unobserved common causes).
50%UnverifiedThe ID algorithm is a complete algorithm for causal effect identification, and each step of its recursive derivation decomposes the problem along the boundaries of c-components.
50%UnverifiedThe more and smaller the c-components, the easier it is for identification algorithms to decompose step by step; if a large number of variables belong to the same c-component, identification often becomes difficult or even infeasible.
50%UnverifiedThe core criterion of the paper is that a safe clustering operation must respect the topological structure of the c-components of the original graph and must not arbitrarily merge variables belonging to different c-components into the same cluster node.
50%