On identifying total effects in the presence of latent variables and selection bias

Zhihong Cai, Manabu Kuroki
Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, PMLR R6:62-69, 2008.

Abstract

Assume that cause-effect relationships between variables can be described as a directed acyclic graph and the corresponding linear structural equation model. We consider the identification problem of total effects in the presence of latent variables and selection bias between a treatment variable and a response variable. Pearl and his colleagues provided the back door criterion, the front door criterion (Pearl, 2000) and the conditional instrumental variable method (Brito and Pearl, 2002) as identifiability criteria for total effects in the presence of latent variables, but not in the presence of selection bias. In order to solve this problem, we propose new graphical identifiability criteria for total effects based on the identifiable factor models. The results of this paper are useful to identify total effects in observational studies and provide a new viewpoint to the identification conditions of factor models.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR6-cai08a, title = {On identifying total effects in the presence of latent variables and selection bias}, author = {Cai, Zhihong and Kuroki, Manabu}, booktitle = {Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence}, pages = {62--69}, year = {2008}, editor = {McAllester, David A. and Myllymäki, Petri}, volume = {R6}, series = {Proceedings of Machine Learning Research}, month = {09--12 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r6/main/assets/cai08a/cai08a.pdf}, url = {https://proceedings.mlr.press/r6/cai08a.html}, abstract = {Assume that cause-effect relationships between variables can be described as a directed acyclic graph and the corresponding linear structural equation model. We consider the identification problem of total effects in the presence of latent variables and selection bias between a treatment variable and a response variable. Pearl and his colleagues provided the back door criterion, the front door criterion (Pearl, 2000) and the conditional instrumental variable method (Brito and Pearl, 2002) as identifiability criteria for total effects in the presence of latent variables, but not in the presence of selection bias. In order to solve this problem, we propose new graphical identifiability criteria for total effects based on the identifiable factor models. The results of this paper are useful to identify total effects in observational studies and provide a new viewpoint to the identification conditions of factor models.}, note = {Reissued by PMLR on 09 October 2024.} }
Endnote
%0 Conference Paper %T On identifying total effects in the presence of latent variables and selection bias %A Zhihong Cai %A Manabu Kuroki %B Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2008 %E David A. McAllester %E Petri Myllymäki %F pmlr-vR6-cai08a %I PMLR %P 62--69 %U https://proceedings.mlr.press/r6/cai08a.html %V R6 %X Assume that cause-effect relationships between variables can be described as a directed acyclic graph and the corresponding linear structural equation model. We consider the identification problem of total effects in the presence of latent variables and selection bias between a treatment variable and a response variable. Pearl and his colleagues provided the back door criterion, the front door criterion (Pearl, 2000) and the conditional instrumental variable method (Brito and Pearl, 2002) as identifiability criteria for total effects in the presence of latent variables, but not in the presence of selection bias. In order to solve this problem, we propose new graphical identifiability criteria for total effects based on the identifiable factor models. The results of this paper are useful to identify total effects in observational studies and provide a new viewpoint to the identification conditions of factor models. %Z Reissued by PMLR on 09 October 2024.
APA
Cai, Z. & Kuroki, M.. (2008). On identifying total effects in the presence of latent variables and selection bias. Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R6:62-69 Available from https://proceedings.mlr.press/r6/cai08a.html. Reissued by PMLR on 09 October 2024.

Related Material