Conditional Common Entropy for Instrumental Variable Testing and Partial Identification

Ziwei Jiang, Murat Kocaoglu
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:21824-21843, 2024.

Abstract

Instrumental variables (IVs) are widely used for estimating causal effects. There are two main challenges when using instrumental variables. First of all, using IV without additional assumptions such as linearity, the causal effect may still not be identifiable. Second, when selecting an IV, the validity of the selected IV is typically not testable since the causal graph is not identifiable from observational data. In this paper, we propose a method for bounding the causal effect with instrumental variables under weak confounding. In addition, we present a novel criterion to falsify the IV with side information about the confounder. We demonstrate the utility of the proposed method with simulated and real-world datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-jiang24b, title = {Conditional Common Entropy for Instrumental Variable Testing and Partial Identification}, author = {Jiang, Ziwei and Kocaoglu, Murat}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {21824--21843}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/jiang24b/jiang24b.pdf}, url = {https://proceedings.mlr.press/v235/jiang24b.html}, abstract = {Instrumental variables (IVs) are widely used for estimating causal effects. There are two main challenges when using instrumental variables. First of all, using IV without additional assumptions such as linearity, the causal effect may still not be identifiable. Second, when selecting an IV, the validity of the selected IV is typically not testable since the causal graph is not identifiable from observational data. In this paper, we propose a method for bounding the causal effect with instrumental variables under weak confounding. In addition, we present a novel criterion to falsify the IV with side information about the confounder. We demonstrate the utility of the proposed method with simulated and real-world datasets.} }
Endnote
%0 Conference Paper %T Conditional Common Entropy for Instrumental Variable Testing and Partial Identification %A Ziwei Jiang %A Murat Kocaoglu %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-jiang24b %I PMLR %P 21824--21843 %U https://proceedings.mlr.press/v235/jiang24b.html %V 235 %X Instrumental variables (IVs) are widely used for estimating causal effects. There are two main challenges when using instrumental variables. First of all, using IV without additional assumptions such as linearity, the causal effect may still not be identifiable. Second, when selecting an IV, the validity of the selected IV is typically not testable since the causal graph is not identifiable from observational data. In this paper, we propose a method for bounding the causal effect with instrumental variables under weak confounding. In addition, we present a novel criterion to falsify the IV with side information about the confounder. We demonstrate the utility of the proposed method with simulated and real-world datasets.
APA
Jiang, Z. & Kocaoglu, M.. (2024). Conditional Common Entropy for Instrumental Variable Testing and Partial Identification. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:21824-21843 Available from https://proceedings.mlr.press/v235/jiang24b.html.

Related Material