Difference-in-Differences Meets Tree-based Methods: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding

Caizhi Tang, Huiyuan Wang, Xinyu Li, Qing Cui, Longfei Li, Jun Zhou
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:33792-33803, 2023.

Abstract

This study considers the estimation of conditional causal effects in the presence of unmeasured confounding for a balanced panel with treatment imposed at the last time point. To address this, we combine Difference-in-differences (DiD) and tree-based methods and propose a new identification assumption that allows for the violation of the (conditional) parallel trends assumption adopted by most existing DiD methods. Under this new assumption, we prove partial identifiability of the conditional average treatment effect on the treated group (CATT). Our proposed method estimates CATT through a tree-based causal approach, guided by a novel splitting rule that avoids model misspecification and unnecessary auxiliary parameter estimation. The splitting rule measures both the error of fitting observed data and the violation of conditional parallel trends simultaneously. We also develop an ensemble of multiple trees via gradient boosting to further enhance performance. Experimental results on both synthetic and real-world datasets validate the effectiveness of our proposed method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-tang23j, title = {Difference-in-Differences Meets Tree-based Methods: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding}, author = {Tang, Caizhi and Wang, Huiyuan and Li, Xinyu and Cui, Qing and Li, Longfei and Zhou, Jun}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {33792--33803}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/tang23j/tang23j.pdf}, url = {https://proceedings.mlr.press/v202/tang23j.html}, abstract = {This study considers the estimation of conditional causal effects in the presence of unmeasured confounding for a balanced panel with treatment imposed at the last time point. To address this, we combine Difference-in-differences (DiD) and tree-based methods and propose a new identification assumption that allows for the violation of the (conditional) parallel trends assumption adopted by most existing DiD methods. Under this new assumption, we prove partial identifiability of the conditional average treatment effect on the treated group (CATT). Our proposed method estimates CATT through a tree-based causal approach, guided by a novel splitting rule that avoids model misspecification and unnecessary auxiliary parameter estimation. The splitting rule measures both the error of fitting observed data and the violation of conditional parallel trends simultaneously. We also develop an ensemble of multiple trees via gradient boosting to further enhance performance. Experimental results on both synthetic and real-world datasets validate the effectiveness of our proposed method.} }
Endnote
%0 Conference Paper %T Difference-in-Differences Meets Tree-based Methods: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding %A Caizhi Tang %A Huiyuan Wang %A Xinyu Li %A Qing Cui %A Longfei Li %A Jun Zhou %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-tang23j %I PMLR %P 33792--33803 %U https://proceedings.mlr.press/v202/tang23j.html %V 202 %X This study considers the estimation of conditional causal effects in the presence of unmeasured confounding for a balanced panel with treatment imposed at the last time point. To address this, we combine Difference-in-differences (DiD) and tree-based methods and propose a new identification assumption that allows for the violation of the (conditional) parallel trends assumption adopted by most existing DiD methods. Under this new assumption, we prove partial identifiability of the conditional average treatment effect on the treated group (CATT). Our proposed method estimates CATT through a tree-based causal approach, guided by a novel splitting rule that avoids model misspecification and unnecessary auxiliary parameter estimation. The splitting rule measures both the error of fitting observed data and the violation of conditional parallel trends simultaneously. We also develop an ensemble of multiple trees via gradient boosting to further enhance performance. Experimental results on both synthetic and real-world datasets validate the effectiveness of our proposed method.
APA
Tang, C., Wang, H., Li, X., Cui, Q., Li, L. & Zhou, J.. (2023). Difference-in-Differences Meets Tree-based Methods: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:33792-33803 Available from https://proceedings.mlr.press/v202/tang23j.html.

Related Material