Continuous Treatment Effects with Surrogate Outcomes

Zhenghao Zeng, David Arbour, Avi Feller, Raghavendra Addanki, Ryan A. Rossi, Ritwik Sinha, Edward Kennedy
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:58306-58328, 2024.

Abstract

In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, we establish the asymptotic normality of the proposed estimator and show possible improvements on the variance compared with methods that solely use labeled data. Extensive simulations show our methods enjoy appealing empirical performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-zeng24a, title = {Continuous Treatment Effects with Surrogate Outcomes}, author = {Zeng, Zhenghao and Arbour, David and Feller, Avi and Addanki, Raghavendra and Rossi, Ryan A. and Sinha, Ritwik and Kennedy, Edward}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {58306--58328}, 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/zeng24a/zeng24a.pdf}, url = {https://proceedings.mlr.press/v235/zeng24a.html}, abstract = {In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, we establish the asymptotic normality of the proposed estimator and show possible improvements on the variance compared with methods that solely use labeled data. Extensive simulations show our methods enjoy appealing empirical performance.} }
Endnote
%0 Conference Paper %T Continuous Treatment Effects with Surrogate Outcomes %A Zhenghao Zeng %A David Arbour %A Avi Feller %A Raghavendra Addanki %A Ryan A. Rossi %A Ritwik Sinha %A Edward Kennedy %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-zeng24a %I PMLR %P 58306--58328 %U https://proceedings.mlr.press/v235/zeng24a.html %V 235 %X In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, we establish the asymptotic normality of the proposed estimator and show possible improvements on the variance compared with methods that solely use labeled data. Extensive simulations show our methods enjoy appealing empirical performance.
APA
Zeng, Z., Arbour, D., Feller, A., Addanki, R., Rossi, R.A., Sinha, R. & Kennedy, E.. (2024). Continuous Treatment Effects with Surrogate Outcomes. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:58306-58328 Available from https://proceedings.mlr.press/v235/zeng24a.html.

Related Material