Multiply-Robust Causal Change Attribution

Victor Quintas-Martinez, Mohammad Taha Bahadori, Eduardo Santiago, Jeff Mu, David Heckerman
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:41821-41840, 2024.

Abstract

Comparing two samples of data, we observe a change in the distribution of an outcome variable. In the presence of multiple explanatory variables, how much of the change can be explained by each possible cause? We develop a new estimation strategy that, given a causal model, combines regression and re-weighting methods to quantify the contribution of each causal mechanism. Our proposed methodology is multiply robust, meaning that it still recovers the target parameter under partial misspecification. We prove that our estimator is consistent and asymptotically normal. Moreover, it can be incorporated into existing frameworks for causal attribution, such as Shapley values, which will inherit the consistency and large-sample distribution properties. Our method demonstrates excellent performance in Monte Carlo simulations, and we show its usefulness in an empirical application. Our method is implemented as part of the Python library “DoWhy“ (Sharma & Kiciman, 2020; Blöbaum et al., 2022).

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-quintas-martinez24a, title = {Multiply-Robust Causal Change Attribution}, author = {Quintas-Martinez, Victor and Bahadori, Mohammad Taha and Santiago, Eduardo and Mu, Jeff and Heckerman, David}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {41821--41840}, 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/quintas-martinez24a/quintas-martinez24a.pdf}, url = {https://proceedings.mlr.press/v235/quintas-martinez24a.html}, abstract = {Comparing two samples of data, we observe a change in the distribution of an outcome variable. In the presence of multiple explanatory variables, how much of the change can be explained by each possible cause? We develop a new estimation strategy that, given a causal model, combines regression and re-weighting methods to quantify the contribution of each causal mechanism. Our proposed methodology is multiply robust, meaning that it still recovers the target parameter under partial misspecification. We prove that our estimator is consistent and asymptotically normal. Moreover, it can be incorporated into existing frameworks for causal attribution, such as Shapley values, which will inherit the consistency and large-sample distribution properties. Our method demonstrates excellent performance in Monte Carlo simulations, and we show its usefulness in an empirical application. Our method is implemented as part of the Python library “DoWhy“ (Sharma & Kiciman, 2020; Blöbaum et al., 2022).} }
Endnote
%0 Conference Paper %T Multiply-Robust Causal Change Attribution %A Victor Quintas-Martinez %A Mohammad Taha Bahadori %A Eduardo Santiago %A Jeff Mu %A David Heckerman %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-quintas-martinez24a %I PMLR %P 41821--41840 %U https://proceedings.mlr.press/v235/quintas-martinez24a.html %V 235 %X Comparing two samples of data, we observe a change in the distribution of an outcome variable. In the presence of multiple explanatory variables, how much of the change can be explained by each possible cause? We develop a new estimation strategy that, given a causal model, combines regression and re-weighting methods to quantify the contribution of each causal mechanism. Our proposed methodology is multiply robust, meaning that it still recovers the target parameter under partial misspecification. We prove that our estimator is consistent and asymptotically normal. Moreover, it can be incorporated into existing frameworks for causal attribution, such as Shapley values, which will inherit the consistency and large-sample distribution properties. Our method demonstrates excellent performance in Monte Carlo simulations, and we show its usefulness in an empirical application. Our method is implemented as part of the Python library “DoWhy“ (Sharma & Kiciman, 2020; Blöbaum et al., 2022).
APA
Quintas-Martinez, V., Bahadori, M.T., Santiago, E., Mu, J. & Heckerman, D.. (2024). Multiply-Robust Causal Change Attribution. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:41821-41840 Available from https://proceedings.mlr.press/v235/quintas-martinez24a.html.

Related Material