Packages & Codes

did_multiplegt_dyn estimates the effect of a treatment on an outcome, using group-level panel data with multiple groups and periods. It computes the DID event-study estimators introduced in de Chaisemartin and D'Haultfoeuille (2020a). The command can be used with a binary and absorbing (staggered) treatment but it can also be used with a non-binary treatment that can increase or decrease multiple times, even if lagged treatments affect the outcome, and if the current and lagged treatments have heterogeneous effects, across space and/or over time. The event-study estimators computed by the command rely on no-anticipation and parallel trends assumptions.

With: Clement de Chaisemartin, Diego Ciccia, Xavier d'Haultfoeuille, Felix Knau, and Melitine Malezieux.

Examples and details

Reference Paper: Difference-in-Differences Estimators of Intertemporal Treatment Effects, (2020). de Chaisemartin, C and D'Haultfoeuille, X.

did_multiplegt_stat estimates difference-in-differences estimators for continuous treatments with heterogeneous effects, assuming that between consecutive periods, the treatment of some units (the "switchers") changes, while the treatment of other units does not. It computes the three estimators (including an IV-related estimator) introduced in de Chaisemartin, C, D'Haultfoeuille, X, Pasquier, F, Sow, D, Vazquez‐Bare, G (2024). The estimators assume static effects and rely on parallel trends assumptions. The command can be used with more than two periods, and for each pair of successive periods, the requested DID estimators (AS, WAS, or IV-WAS) are computed, as well as their aggregated versions defined as a weighted average of the estimators of the different periods. It also supports unbalanced panel data and doubly-robust estimators, using cross-validation.

With: Clement de Chaisemartin, Diego Ciccia, Xavier d'Haultfoeuille, Felix Knau, Felix Pasquier and Gonzalo Vasquez-Bare.

Examples and details

Reference Paper: Difference-in-Differences for Continuous Treatments and Instruments with Stayers. (2024). de Chaisemartin, C, D'Haultfoeuille, X, Pasquier, F, Sow, D, Vazquez‐Bare, G.

did_had estimates the effect of a treatment on an outcome in a heterogeneous adoption design (HAD) with no stayers but some quasi-stayers. HADs are designs where all groups are untreated in the first period, and then some groups receive a strictly positive treatment dose at a later period F, which is the same for all treated groups (with variation in treatment timing, the did_multiplegt_dyn package may be used).

With: Clement de Chaisemartin, Diego Ciccia, Xavier d'Haultfoeuille, and Felix Knau.

Examples and details

Reference Paper:  Two-way Fixed Effects and Differences-in-Differences in Heterogeneous Adoption Designs without Stayers. (2024). Clement de Chaisemartin, Diego Ciccia, Xavier d'Haultfoeuille, and Felix Knau.

stute_test implements the non-parametric test that the expectation of Y given D is linear proposed by Stute (1997). In the companion vignette, we sketch the intuition behind the test, as to motivate the use of the package and its options. Please refer to Stute (1997) and Section 3 of de Chaisemartin and D'Haultfoeuille (2024) for further details.

yatchew_test implements a non-parametric test that the expectation of Y given D is linear. The program implements both the original test proposed by Yatchew (1997) and its heteroskedasticity-robust version proposed by de Chaisemartin, D'Haultfoeuille \& Gurgand (2024)

With: Clement de Chaisemartin, Diego Ciccia, Xavier d'Haultfoeuille, and Felix Knau

Examples and details