cv_approximate_fs: CV for approximate forward stepwise selection

In stephenbates19/logratiolasso: Tools For Large Log Ratio Models

Description

Determines the number of steps of approximate forward stepwise selection by cross-validation.

Usage

cv_approximate_fs(x, y, k_max = 5, n_folds = 10, fold_id = NULL, ...)

Arguments

x	The input matrix. Assumed to be the logarithmic transform of positive raw inputs.
y	The real-valued response.
k_max	The largest number of log-ratios to consider.
n_folds	The number of folds. Default is 10.
fold_id	Optional fold assignments. Default is uniformly random assignment.

Value

cvm The CV estimate of prediction error for each model size.

beta The coefficient vector corresponding to the best model size chosen by CV.

intercept. The intercept of the best model chosen by CV.

model A approximate_fs object with k_max set to by the optimal model size chose by CV.

stephenbates19/logratiolasso documentation built on May 18, 2019, 4:52 p.m.