linearmodels.panel.results.RandomEffectsResults.predict¶

In- and out-of-sample predictions.

Parameters:¶

exog: PanelData | ndarray | DataArray | DataFrame | Series | None = None¶: Exogenous values to use in out-of-sample prediction (nobs by nexog). The containers are those of exog in the model. A list or a tuple raises a TypeError that names exog. If neither exog nor data is provided, the in-sample values are returned.
data: DataFrame | None = None¶: DataFrame to use for out-of-sample predictions when model was constructed using a formula.
fitted: bool = True¶: Flag indicating whether to include the fitted values. The default is True. Only used for in-sample values.
effects: bool = False¶: Flag indicating whether to include estimated effects. The default is False. Only used for in-sample values.
idiosyncratic: bool = False¶: Flag indicating whether to include the estimated idiosyncratic shock. The default is False. Only used for in-sample values.
missing: bool = False¶: Flag indicating to adjust for dropped observations. If True, the values returned will have the same size as the original input data before filtering missing values. The default is False.

Returns:¶

DataFrame containing columns for all selected output.

Return type:¶

pandas.DataFrame

Raises:¶

ValueError – If both exog and data are provided, if no in-sample output is selected, or if effects or idiosyncratic is requested for out-of-sample predictions.

Notes

data can only be used when the model was created using the formula interface. exog can be used for both a model created using a formula or a model specified with dependent and exog arrays.

When using exog to generate out-of-sample predictions, the variable order must match the variables in the original model.

Idiosyncratic errors and effects are not available for out-of-sample predictions.