SyntheticDID

Guide & worked example · Source

SyntheticDID(
    eta_omega=None,
    eta_lambda=1e-06,
    zeta_omega=None,
    zeta_lambda=None,
    omega_intercept=True,
    lambda_intercept=True,
    min_decrease=1e-08,
    maxiter=1000,
    sparsify=True,
    omega=None,
    lambda_weights=None,
    update_omega=None,
    update_lambda=None,
    device=None,
)

Synthetic difference-in-differences with two-way balancing weights.

Attributes

Name	Description
`device`
`eta_lambda`
`eta_omega`
`lambda_intercept`
`lambda_weights`
`maxiter`
`min_decrease`
`omega`
`omega_intercept`
`params`
`sparsify`
`update_lambda`
`update_omega`
`zeta_lambda`
`zeta_omega`

Methods

Name	Description
fit
predict
sparsify_weights
summary	Print a summary of the model results.
to	Move all model parameters to specified device.

fit

SyntheticDID.fit(Y, N0, T0)

predict

SyntheticDID.predict()

sparsify_weights

SyntheticDID.sparsify_weights(weights)

summary

SyntheticDID.summary()

Print a summary of the model results.

to

SyntheticDID.to(device)

Move all model parameters to specified device.

Parameters

Name	Type	Description	Default
device	torch.device or str	Target device (‘cuda’, ‘cpu’, etc.)	required

Returns

Name	Type	Description
	BaseEstimator	Self for method chaining