论文信息 - Identifiability of Path-Specific Effects

Identifiability of Path-Specific Effects

Counterfactual quantities representing path-specific effects arise in cases where we are interested in computing the effect of one variable on another only along certain causal paths in the graph (in other words by excluding a set of edges from consideration). A recent paper [Pearl, 2001] details a method by which such an exclusion can be specified formally by fixing the value of the parent node of each excluded edge. In this paper we derive simple, graphical conditions for experimental identifiability of path-specific effects, namely, conditions under which path-specific effects can be estimated consistently from data obtained from controlled experiments.

[1] Judea Pearl,et al. Direct and Indirect Effects , 2001, UAI.

[2] Joseph Y. Halpern. Axiomatizing Causal Reasoning , 1998, UAI.

[3] J. Pearl. Causality: Models, Reasoning and Inference , 2000 .

[4] Nils J. Nilsson,et al. Artificial Intelligence , 1974, IFIP Congress.

[5] Arne Collen. The Foundation of Science , 1997 .

[6] J. Robins,et al. Identifiability and Exchangeability for Direct and Indirect Effects , 1992, Epidemiology.

[7] J. Pearl,et al. An Axiomatic Characterization of Causal Counterfactuals , 1998 .

[8] Judea Pearl,et al. Testing Identifiability of Causal Effects , 1995, UAI.

[9] Louette R. Johnson Lutjens. Research , 2006 .

[10] James M. Robins,et al. Causal Inference from Complex Longitudinal Data , 1997 .

[11] Judea Pearl,et al. Counterfactual Probabilities: Computational Methods, Bounds and Applications , 1994, UAI.

[12] Joseph Y. Halpern. An Analysis of First-Order Logics of Probability , 1989, IJCAI.

[13] D. Rubin. Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .