Exploiting Equality Constraints in Causal Inference

Chi Zhang, Carlos Cinelli, Bryant Chen, Judea Pearl
Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, PMLR 130:1630-1638, 2021.

Abstract

Assumptions about equality of effects are commonly made in causal inference tasks. For example, the well-known “difference-in-differences” method assumes that confounding remains constant across time periods. Similarly, it is not unreasonable to assume that causal effects apply equally to units undergoing interference. Finally, sensitivity analysis often hypothesizes equality among existing and unaccounted for confounders. Despite the ubiquity of these “equality constraints,” modern identification methods have not leveraged their presence in a systematic way. In this paper, we develop a novel graphical criterion that extends the well-known method of generalized instrumental sets to exploit such additional constraints for causal identification in linear models. We further demonstrate how it solves many diverse problems found in the literature in a general way, including difference-in-differences, interference, as well as benchmarking in sensitivity analysis.

Cite this Paper

BibTeX
@InProceedings{pmlr-v130-zhang21c, title = { Exploiting Equality Constraints in Causal Inference }, author = {Zhang, Chi and Cinelli, Carlos and Chen, Bryant and Pearl, Judea}, booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics}, pages = {1630--1638}, year = {2021}, editor = {Banerjee, Arindam and Fukumizu, Kenji}, volume = {130}, series = {Proceedings of Machine Learning Research}, month = {13--15 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v130/zhang21c/zhang21c.pdf}, url = {https://proceedings.mlr.press/v130/zhang21c.html}, abstract = { Assumptions about equality of effects are commonly made in causal inference tasks. For example, the well-known “difference-in-differences” method assumes that confounding remains constant across time periods. Similarly, it is not unreasonable to assume that causal effects apply equally to units undergoing interference. Finally, sensitivity analysis often hypothesizes equality among existing and unaccounted for confounders. Despite the ubiquity of these “equality constraints,” modern identification methods have not leveraged their presence in a systematic way. In this paper, we develop a novel graphical criterion that extends the well-known method of generalized instrumental sets to exploit such additional constraints for causal identification in linear models. We further demonstrate how it solves many diverse problems found in the literature in a general way, including difference-in-differences, interference, as well as benchmarking in sensitivity analysis. } }
Endnote
%0 Conference Paper %T Exploiting Equality Constraints in Causal Inference %A Chi Zhang %A Carlos Cinelli %A Bryant Chen %A Judea Pearl %B Proceedings of The 24th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2021 %E Arindam Banerjee %E Kenji Fukumizu %F pmlr-v130-zhang21c %I PMLR %P 1630--1638 %U https://proceedings.mlr.press/v130/zhang21c.html %V 130 %X Assumptions about equality of effects are commonly made in causal inference tasks. For example, the well-known “difference-in-differences” method assumes that confounding remains constant across time periods. Similarly, it is not unreasonable to assume that causal effects apply equally to units undergoing interference. Finally, sensitivity analysis often hypothesizes equality among existing and unaccounted for confounders. Despite the ubiquity of these “equality constraints,” modern identification methods have not leveraged their presence in a systematic way. In this paper, we develop a novel graphical criterion that extends the well-known method of generalized instrumental sets to exploit such additional constraints for causal identification in linear models. We further demonstrate how it solves many diverse problems found in the literature in a general way, including difference-in-differences, interference, as well as benchmarking in sensitivity analysis.
APA
Zhang, C., Cinelli, C., Chen, B. & Pearl, J.. (2021). Exploiting Equality Constraints in Causal Inference . Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 130:1630-1638 Available from https://proceedings.mlr.press/v130/zhang21c.html.

Related Material