statsmodels.tsa.ardl.ARDLResults.apply

ARDLResults.apply(endog, exog=None, fixed=None, refit=None, fit_kwargs=None)

Apply the fitted parameters to new data unrelated to original data. It creates a new result object by using the current fitted parameters. Applied to a completely new dataset, assumed to be unrelated to the model’s original data.

The new results can then be used for analysis or forecasting.

Parameters:

endogarray_like

New observations from the modeled time-series process.

exogarray_like, optional

New observations of exogenous regressors, if applicable. It must supply the same un-lagged columns as the original exog. The lag structure fitting the original data is re-applied automatically.

fixedarray_like, optional

New observations of fixed regressors, if applicable.

refitbool, optional

Whether to re-fit the parameters, using the new dataset. Default is False (so, the parameters from the current results object are used to create the new results object.)

fit_kwargsdict, optional

Keyword arguments to pass to fit (if refit=True)

Returns:

ARDLResults

Updated results object containing results for the new dataset.

See also

ARDLResults.append

statsmodels.tsa.ar_model.AutoRegResults.apply

Notes

The endog argument given to this method should consist of new observations that are not necessarily related to the original model’s endog dataset.

One should be careful when using deterministic processes with cyclical components such as seasonal dummies or Fourier series. These deterministic components will align to the first observation in the data and so it is necessary that any new data should have the same initial period.