ensemble: This function implements an ensemble learner with the...
In MCKnaus/causalDML: Causal Double Machine Learning

ensemble

R Documentation

This function implements an ensemble learner with the possibility to extract the smoother matrix of linear smoothers.

This function implements an ensemble learner with the possibility to extract the smoother matrix of linear smoothers.

ensemble(ml, x, y, xnew, nfolds = 5, weights = FALSE, quiet = TRUE)

`ml`	List of methods build via `create_method` to be used in ensemble.
`x`	Covariate matrix of training sample.
`y`	Vector of outcomes of training sample.
`xnew`	Covariate matrix of test sample.
`nfolds`	Number of folds used in cross-validation of ensemble weights (default `nfolds=5`).
`weights`	If TRUE, weights underlying the ensemble prediction for xnew calculated.
`quiet`	If FALSE, print method that is currently computed.

Ensemble object containing:

`ensemble`	the predictions of the trained ensemble for the test sample
`best`	the predictions of the best performing single method for the test sample
`fit_full`	matrix containing predictions of all considered methods
`weights`	If `weights=TRUE`, smoother matrix of dimension `nrow(xnew)` x `nrow(x)` containing the weights that deliver ensemble predictions where each row gives the weight that each training outcome received in the prediction
`nnls_weights`	the weights that each method receives in the ensemble
`mse_cv`	cross-validated MSEs for each method
`fit_cv`	matrix containing the cross-validated predictions for each method