Time-Aware Synthetic Control

Saeyoung Rho, Cyrus Illick, Samhitha Narasipura, Alberto Abadie, Daniel Hsu, Vishal Misra
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:2179-2187, 2026.

Abstract

The synthetic control (SC) framework is widely used for observational causal inference with time-series panel data. Despite its success across diverse applications, existing SC methods typically treat pre-intervention time indices as exchangeable, meaning they may fail to exploit temporal structure when strong trends are present. We propose Time-Aware Synthetic Control (TASC), a method that addresses this limitation by adopting a state-space model with a constant trend component while preserving the low-rank structure of the signal. TASC uses the Kalman filter and the Rauch–Tung–Striebel smoother in two steps: it first fits a generative time-series model with expectation–maximization and then performs counterfactual inference. We evaluate TASC on simulated and real-world datasets spanning policy evaluation and sports prediction. Our results demonstrate that TASC offers advantages in settings with high observation noise and long prediction horizons.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-rho26a, title = { Time-Aware Synthetic Control }, author = {Rho, Saeyoung and Illick, Cyrus and Narasipura, Samhitha and Abadie, Alberto and Hsu, Daniel and Misra, Vishal}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {2179--2187}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/rho26a/rho26a.pdf}, url = {https://proceedings.mlr.press/v300/rho26a.html}, abstract = { The synthetic control (SC) framework is widely used for observational causal inference with time-series panel data. Despite its success across diverse applications, existing SC methods typically treat pre-intervention time indices as exchangeable, meaning they may fail to exploit temporal structure when strong trends are present. We propose Time-Aware Synthetic Control (TASC), a method that addresses this limitation by adopting a state-space model with a constant trend component while preserving the low-rank structure of the signal. TASC uses the Kalman filter and the Rauch–Tung–Striebel smoother in two steps: it first fits a generative time-series model with expectation–maximization and then performs counterfactual inference. We evaluate TASC on simulated and real-world datasets spanning policy evaluation and sports prediction. Our results demonstrate that TASC offers advantages in settings with high observation noise and long prediction horizons. } }
Endnote
%0 Conference Paper %T Time-Aware Synthetic Control %A Saeyoung Rho %A Cyrus Illick %A Samhitha Narasipura %A Alberto Abadie %A Daniel Hsu %A Vishal Misra %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-rho26a %I PMLR %P 2179--2187 %U https://proceedings.mlr.press/v300/rho26a.html %V 300 %X The synthetic control (SC) framework is widely used for observational causal inference with time-series panel data. Despite its success across diverse applications, existing SC methods typically treat pre-intervention time indices as exchangeable, meaning they may fail to exploit temporal structure when strong trends are present. We propose Time-Aware Synthetic Control (TASC), a method that addresses this limitation by adopting a state-space model with a constant trend component while preserving the low-rank structure of the signal. TASC uses the Kalman filter and the Rauch–Tung–Striebel smoother in two steps: it first fits a generative time-series model with expectation–maximization and then performs counterfactual inference. We evaluate TASC on simulated and real-world datasets spanning policy evaluation and sports prediction. Our results demonstrate that TASC offers advantages in settings with high observation noise and long prediction horizons.
APA
Rho, S., Illick, C., Narasipura, S., Abadie, A., Hsu, D. & Misra, V.. (2026). Time-Aware Synthetic Control . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:2179-2187 Available from https://proceedings.mlr.press/v300/rho26a.html.

Related Material