Efficiently Learning Synthetic Control Models for High-dimensional Disaggregated Data

2025-10-26

AI Summary

To address the computational inefficiency and implementation challenges of Synthetic Control (SC) in high-dimensional, fine-grained data across multiple treated units, this paper introduces the multivariate square-root Lasso into the SC framework—its first application to SC. The proposed method jointly performs high-dimensional covariate selection and models temporal dependence structures in outcomes. Theoretically, we derive a novel error bound tailored to this setting, enabling the first precise quantification of estimation error for the Average Treatment Effect on the Treated (ATT). Methodologically, the approach substantially improves both computational efficiency and robustness to heteroskedasticity and serial correlation. Empirically, we apply it to estimate the causal effect of county-level stay-at-home orders on unemployment rates in the U.S., accurately uncovering heterogeneous policy impacts. This work extends the applicability of SC to high-dimensional, multi-unit treatment settings and delivers a scalable, interpretable tool for causal policy evaluation.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

The Synthetic Control method (SC) has become a valuable tool for estimating causal effects. Originally designed for single-treated unit scenarios, it has recently found applications in high-dimensional disaggregated settings with multiple treated units. However, challenges in practical implementation and computational efficiency arise in such scenarios. To tackle these challenges, we propose a novel approach that integrates the Multivariate Square-root Lasso method into the synthetic control framework. We rigorously establish the estimation error bounds for fitting the Synthetic Control weights using Multivariate Square-root Lasso, accommodating high-dimensionality and time series dependencies. Additionally, we quantify the estimation error for the Average Treatment Effect on the Treated (ATT). Through simulation studies, we demonstrate that our method offers superior computational efficiency without compromising estimation accuracy. We apply our method to assess the causal impact of COVID-19 Stay-at-Home Orders on the monthly unemployment rate in the United States at the county level.

Problem

• Extending Synthetic Control method for high-dimensional disaggregated data
• Addressing computational challenges in multi-unit causal estimation
• Quantifying treatment effects with improved efficiency and accuracy

Innovation

• Integrates Multivariate Square-root Lasso into synthetic control
• Establishes error bounds for high-dimensional time series data
• Provides computational efficiency without sacrificing estimation accuracy