macrosynergy.learning.forecasting.bootstrap.base_modified_regressor#

class BaseModifiedRegressor(model, method, error_offset=1e-05, bootstrap_method='panel', bootstrap_iters=100, resample_ratio=1, max_features=None, analytic_method=None)[source]#

Bases: BaseEstimator, RegressorMixin, BasePanelBootstrap, ABC

fit(X, y)[source]#

Fit a linear model and modify coefficients based on standard errors.

Parameters:

  • X (pd.DataFrame) – Input feature matrix.

  • y (pd.DataFrame or pd.Series) – Target vector associated with each sample in X.

Returns:

Fitted estimator.

Return type:

self

predict(X)[source]#

Predict using the unadjusted linear model.

Parameters:

X (pd.DataFrame) – Input feature matrix.

Returns:

Predicted values.

Return type:

np.ndarray or pd.Series

create_signal(X)[source]#

Predict using the coefficient-adjusted linear model.

Parameters:

X (pd.DataFrame) – Input feature matrix.

Returns:

Signal from the adjusted factor model based on X.

Return type:

np.ndarray or pd.Series

Notes

We define an additional create_signal method instead of using the predict method in order to not interfere with hyperparameter searches with standard metrics. Moreover, outputs from the adjusted factor model are not valid predictions, but are valid trading signals.

adjust_bootstrap_se(model, X, y)[source]#

Adjust the coefficients of the linear model by bootstrap standard errors.

Parameters:

  • model (RegressorMixin) – The underlying linear model to be modified.

  • X (pd.DataFrame) – Input feature matrix.

  • y (pd.DataFrame or pd.Series) – Target vector associated with each sample in X.

Returns:

  • intercept (float) – Adjusted intercept.

  • coef (np.ndarray) – Adjusted coefficients.

adjust_analytical_se(model, X, y, analytic_method)[source]#

Adjust the coefficients of the linear model by an analytical standard error formula.

Parameters:

  • model (RegressorMixin) – The underlying linear model to be modified.

  • X (pd.DataFrame) – Input feature matrix.

  • y (pd.DataFrame or pd.Series) – Target vector associated with each sample in X.

  • analytic_method (str) – The analytic method used to calculate standard errors.

Returns:

  • intercept (float) – Adjusted intercept.

  • coef (np.ndarray) – Adjusted coefficients.

Notes

Analytical standard errors are model-specific, meaning that they must be implemented in a subclass of BaseModifiedRegressor.

set_score_request(*, sample_weight: bool | None | str = '$UNCHANGED$') → BaseModifiedRegressor#

Request metadata passed to the score method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to score.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

sample_weight (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_weight parameter in score.

Returns:

self – The updated object.

Return type:

object