moderndid.EtwfeResult#

class moderndid.EtwfeResult(coefficients: ndarray, std_errors: ndarray, vcov: ndarray, coef_names: list, gt_pairs: list, n_obs: int, n_units: int, r_squared: float | None = None, adj_r_squared: float | None = None, data: object = None, config: object = None, estimation_params: dict = {})[source]#

Bases: NamedTuple

Container for ETWFE regression output.

This class implements the maketables plug-in interface for publication-quality tables. See Publication Tables with maketables.

Returned by etwfe. Stores the saturated TWFE regression coefficients and variance-covariance matrix needed by emfx for aggregation.

Attributes:

coefficientsnumpy.ndarray: Coefficient estimates for each cohort x time interaction term.
std_errorsnumpy.ndarray: Standard errors for each coefficient.
vcovnumpy.ndarray: Variance-covariance matrix of the interaction coefficients.
coef_nameslist[str]: Names for each coefficient (from pyfixest).
gt_pairslist[tuple[float, float]]: (group, time) pair for each coefficient.
n_obsint: Number of observations used in estimation.
n_unitsint: Number of unique cross-sectional units.
r_squaredfloat or None: R-squared of the regression.
adj_r_squaredfloat or None: Adjusted R-squared of the regression.
dataobject or None: Preprocessed DataFrame (used by emfx for cell counts).
configobject or None: EtwfeConfig used for estimation.
estimation_paramsdict: Additional estimation parameters.

Methods

`count`(value, /)	Return number of occurrences of value.
`index`(value[, start, stop])	Return first index of value.

Attributes

`adj_r_squared`	Adjusted R-squared of the regression.
`coef_names`	Names for each coefficient from pyfixest.
`coefficients`	Coefficient estimates for each cohort x time interaction term.
`config`	EtwfeConfig used for estimation.
`data`	Preprocessed DataFrame (used by emfx for aggregation).
`estimation_params`	Estimation parameters (yname, cgroup, formula, etc.).
`gt_pairs`	(group, time) pair for each coefficient.
`n_obs`	Number of observations used in estimation.
`n_units`	Number of unique cross-sectional units.
`r_squared`	R-squared of the regression.
`std_errors`	Standard errors for each coefficient.
`vcov`	Variance-covariance matrix of the interaction coefficients.