Munich Personal RePEc Archive

Revisiting the Synthetic Control Estimator

Ferman, Bruno and Pinto, Cristine (2016): Revisiting the Synthetic Control Estimator.

This is the latest version of this item.

[img]
Preview
PDF
MPRA_paper_86495.pdf

Download (956kB) | Preview

Abstract

We analyze the conditions under which the Synthetic Control (SC) estimator is unbiased. We show that the SC estimator is generally biased if treatment assignment is correlated with unobserved confounders, even when the number of pre-treatment periods goes to infinity, and in settings where one should expect an almost perfect pre-treatment fit. While our results suggest that researchers should be more careful in interpreting the identification assumptions required for the SC method, we show that, with a slight modification, the SC method can substantially improve in terms of bias and variance relative to standard methods.

Available Versions of this Item

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.