k_var_model: Best subset selection for linear regression

In iBART: Iterative Bayesian Additive Regression Trees Descriptor Selection Method

Best subset selection for linear regression

Description

Best subset selection for linear regression

Usage

k_var_model(
  X_train,
  y_train,
  X_test = NULL,
  y_test = NULL,
  k = 1,
  parallel = FALSE
)

Arguments

X_train	The design matrix used during training.
y_train	The response variable used during training.
X_test	The design matrix used during testing. Default is X_test = NULL and full data will be used to train the best subset linear regression model.
y_test	The response variable used during testing. Default is y_test = NULL and full data will be used to train the best subset linear regression model.
k	The maximum number of predictors allowed in the model. For example, k = 5 will produce the best model 5 predictors.
parallel	Logical flag for parallelization. Default is parallel = FALSE.

Value

A list of outputs.

models	An lm object storing the best k-predictor linear model.
names	The variable name of the best k predictors.
rmse_in	In-sample RMSE of the model.
rmse_out	Out-of-sample RMSE of the model.

iBART documentation built on Nov. 15, 2023, 1:09 a.m.