Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects

By C. Chaisemartin, X. d'Haultfoeuille. The American Economic Review, 2018

3,809 citations recorded by the source corpus.

Linear regressions with period and group fixed effects are widely used to estimate treatment effects. We show that they estimate weighted sums of the average treatment effects (ATE ) in each group and period, with weights that may be negative. Due to the negative weights, the linear regression coefficient may for instance be negative while all the ATEs are positive. We propose another estimator that solves this issue. In the two applications we revisit, it is significantly different from the linear regression estimator. (JEL C21, C23, D72, J31, J51, L82)

Subfields: Economics and Econometrics, Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/f2b059a83a75bd2f8fe4b4f29f53b865eedc024f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Difference-in-Differences with a Continuous Treatment (2021, neutral): “[13] Clement de Chaisemartin and Xavier D’Haultfœuille.”
Difference-in-Differences with multiple time periods (2018, background): “…cannot separately identity the effect of ci and Xi on Yit. our proposal differs from Wooldridge (2005), Chernozhukov et al. (2013), Borusyak and Jaravel (2017), S loczyński (2017), de Chaisemartin and D’Haultfœuille (2018), Goodman-Bacon (2018), Abraham and Sun (2018), and Athey and Imbens (2018).”
Revisiting event study designs: robust and efficient estimation (2021, methodology): “Finally, we compare the efficiency of the imputation estimator to the alternative robust estimators of de Chaisemartin and D’Haultfœuille (2022) and Sun and Abraham (2021), abbreviated dCDH and SA. We document the in-sample efficiency gains in Figure 6 by showing the point estimates (39)We obtain th”
Synthetic Difference In Differences (2018, neutral): “The model allows for heterogeneity in treatment effects τij, as in de Chaisemartin and d’Haultfoeuille [2019]. As above, we assume block assignment Wit = 1 ({i > Nco, t > Tpre}), where the subscript “co” stands for control group, “tr” stands for treatment group, “pre” stands for pre-treatment, and “”
Difference-in-Differences Estimators of Intertemporal Treatment Effects (2020, positive): “Schmidheiny and Siegloch (2020) have shown that the two event-study regressions are numerically identical up to a linear bijection.”
Two-way Fixed Effects and Differences-in-Differences Estimators with Several Treatments (2020, methodology): “DIDfM extends the DIDM estimator in de Chaisemartin and D’Haultfœuille (2020) to settings with several treatments.”