moderndid.MPResult

class moderndid.MPResult(groups: ndarray, times: ndarray, att_gt: ndarray, vcov_analytical: ndarray, se_gt: ndarray, critical_value: float, influence_func: ndarray, n_units: int | None = None, wald_stat: float | None = None, wald_pvalue: float | None = None, aggregate_effects: object | None = None, alpha: float = 0.05, estimation_params: dict = {}, G: ndarray | None = None, weights_ind: ndarray | None = None)

Bases: NamedTuple

Container for group-time average treatment effect results.

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

Attributes:

groupsnumpy.ndarray

Which group (defined by period first treated) each group-time ATT is for.

timesnumpy.ndarray

Which time period each group-time ATT is for.

att_gtnumpy.ndarray

The group-time average treatment effects for each group-time combination.

vcov_analyticalnumpy.ndarray

Analytical estimator for the asymptotic variance-covariance matrix.

se_gtnumpy.ndarray

Standard errors for group-time ATTs. If bootstrap used, provides bootstrap-based SE.

critical_valuefloat

Critical value - simultaneous if obtaining simultaneous confidence bands, otherwise based on pointwise normal approximation.

influence_funcnumpy.ndarray

The influence function for estimating group-time average treatment effects.

n_unitsint, optional

The number of unique cross-sectional units.

wald_statfloat, optional

The Wald statistic for pre-testing the common trends assumption.

wald_pvaluefloat, optional

The p-value of the Wald statistic for pre-testing common trends.

aggregate_effectsobject, optional

An aggregate treatment effects object.

alphafloat

The significance level (default 0.05).

estimation_paramsdict

Dictionary containing DID estimation parameters including:

• call_info: original function call information

• control_group: ‘nevertreated’ or ‘notyettreated’

• anticipation_periods: number of anticipation periods

• estimation_method: estimation method used

• bootstrap: whether bootstrap was used

• uniform_bands: whether simultaneous confidence bands were computed

• G: unit-level group assignments

• weights_ind: unit-level sampling weights

Methods

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

Attributes

G	Unit-level group assignments.
aggregate_effects	Aggregate treatment effects object.
alpha	Significance level.
att_gt	Group-time average treatment effects.
critical_value	Critical value for confidence intervals.
estimation_params	DID estimation parameters.
groups	Which group (defined by period first treated) each group-time ATT is for.
influence_func	Influence function for estimating group-time average treatment effects.
n_units	Number of unique cross-sectional units.
se_gt	Standard errors for group-time ATTs.
times	Which time period each group-time ATT is for.
vcov_analytical	Analytical estimator for the asymptotic variance-covariance matrix.
wald_pvalue	P-value of the Wald statistic for pre-testing common trends.
wald_stat	Wald statistic for pre-testing common trends.
weights_ind	Unit-level sampling weights.