PValueDropRegressor

class pyoptex.analysis.estimators.p_value_drop_model.PValueDropRegressor(factors=(), Y2X=<function identityY2X>, random_effects=(), conditional=False, threshold=0.05, dependencies=None, mode=None)

A regressor implementing the RegressionMixin and ConditionalRegressionMixin interfaces.

Permits to fit a simple model provided in Y2X with optionally random effects. Model selection from the model matrix is done by dropping terms one-by-one based on the highest p-value, higher than the threshold.

If desired, the model can force weak and strong heredity by setting the mode to ‘weak’ or ‘strong’ respectively, and providing a dependency matrix.

Note

It also includes all parameters and attributes from RegressionMixin and ConditionalRegressionMixin

Attributes

thresholdfloat

The p-value threshold, any term above is dropped (while adhering to the selected mode).

dependenciesnp.array(2d)

The dependency matrix of size (N, N) with N the number of terms in the encoded model (output from Y2X). Term i depends on term j if dep(i, j) = true.

modeNone, ‘weak’ or ‘strong’

The heredity mode to adhere to.

__init__(factors=(), Y2X=<function identityY2X>, random_effects=(), conditional=False, threshold=0.05, dependencies=None, mode=None)

P-value based model selection regressor.

Parameters

factorslist(Factor)

A list of factors to be used during fitting. It contains the categorical encoding, continuous normalization, etc.

Y2Xfunc(Y)

The function to transform a design matrix Y to a model matrix X.

random_effectslist(str)

The names of any random effect columns. Every random effect is interpreted as a string column and encoded using effect encoding.

conditionalbool

Whether to create a conditional model or not.

thresholdfloat

The p-value threshold, any term above is dropped (while adhering to the selected mode).

dependenciesnp.array(2d)

The dependency matrix of size (N, N) with N the number of terms in the encoded model (output from Y2X). Term i depends on term j if dep(i, j) = true.

modeNone, ‘weak’ or ‘strong’

The heredity mode to adhere to.

Methods

PValueDropRegressor.fit(X, y)	Fits the data.
PValueDropRegressor.formula([labels])	Creates the prediction formula of the fit for the encoded and normalized data.
PValueDropRegressor.get_metadata_routing()	Get metadata routing of this object.
PValueDropRegressor.get_params([deep])	Get parameters for this estimator.
PValueDropRegressor.model_formula(model)	Creates the prediction formula of the fit for the encoded and normalized data.
PValueDropRegressor.pred_var(X)	Prediction variances for the new values specified in X.
PValueDropRegressor.predict(X)	Predict on new data after fitting.
PValueDropRegressor.preprocess_fit(X, y)	Preprocesses before fitting the data.
PValueDropRegressor.preprocess_predict(X)	Preprocessing the incoming data before prediction.
PValueDropRegressor.score(X, y[, sample_weight])	Return the coefficient of determination of the prediction.
PValueDropRegressor.set_params(**params)	Set the parameters of this estimator.
PValueDropRegressor.set_score_request(*[, ...])	Request metadata passed to the score method.
PValueDropRegressor.summary()	Generates a summary of the fit in case it was stored during training in the fit_ attribute.

Attributes

PValueDropRegressor.M_	Alias for information_matrix
PValueDropRegressor.Minv_	Alias for inv_information_matrix
PValueDropRegressor.V_	Alias for obs_cov
PValueDropRegressor.Vinv_	Alias for inv_obs_cov
PValueDropRegressor.information_matrix	The information matrix of the fitted data.
PValueDropRegressor.inv_information_matrix	The inverse of the information matrix.
PValueDropRegressor.inv_obs_cov	The inverse of the observation covariance matrix.
PValueDropRegressor.is_fitted	Checks whether the regressor has been fitted.
PValueDropRegressor.obs_cov	The observation covariance matrix \(V = var(Y)\).
PValueDropRegressor.total_var	The total variance on the normalized y-values.