moderndid.twfe_did_rc#

moderndid.twfe_did_rc(y, post, d, covariates=None, i_weights=None, boot=False, boot_type='weighted', nboot=999, influence_func=False)[source]#

Compute linear two-way fixed effects DiD estimator for the ATT with repeated cross-sections.

Implements the linear two-way fixed effects (TWFE) estimator for the ATT with repeated cross-section data, as illustrated in [1]. The estimator is based on the regression model from equation (2.5) of [1] as

\[Y_{it} = \alpha_1 + \alpha_2 T_i + \alpha_3 D_i + \tau^{fe}(T_i \cdot D_i) + \theta' X_i + \varepsilon_{it}.\]

Parameters:

ynumpy.ndarray: A 1D array of outcomes from both pre and post-treatment periods.
postnumpy.ndarray: A 1D array of post-treatment dummies (1 if observation belongs to post-treatment period, 0 if observation belongs to pre-treatment period).
dnumpy.ndarray: A 1D array of group indicators (1 if observation is treated in the post-treatment period, 0 otherwise).
covariatesnumpy.ndarray, optional: A 2D array of covariates to be used in the regression estimation. We will always include an intercept.
i_weightsnumpy.ndarray, optional: A 1D array of weights. If None, then every observation has equal weight. Weights are normalized to have mean 1.
bootbool, default=False: Whether to compute bootstrap standard errors.
boot_type{“weighted”, “multiplier”}, default=”weighted”: Type of bootstrap to be performed (not relevant if boot = False).
nbootint, default=999: Number of bootstrap repetitions (not relevant if boot = False).
influence_funcbool, default=False: Whether to return the influence function.

Returns:

TWFEDIDRCResult: A NamedTuple containing the TWFE DiD point estimate, standard error, confidence interval, bootstrap draws, and influence function.

Warning

This estimator generally does not recover the ATT. We encourage users to adopt alternative specifications.