cv_grid_lasso_logistic: Computes cross-validated solutions for binary response model

In bgs25/SubsetGridRegression: Subset Grid Regression for Lasso, Ridge and Logistic Lasso Regression Models with Potential Prior Information

Description

Cross-validation wrapper for grid_lasso_logistic that computes solutions, selects and fits the optimal model.

Usage

cv_grid_lasso_logistic(
  x = NULL,
  y,
  K = 5,
  var_order = NULL,
  lambda = NULL,
  nlambda = 100L,
  grid.size = p,
  lambda.min.ratio = ifelse(n < p, 0.01, 1e-04),
  thresh = 1e-10,
  maxit = 1e+05,
  mc.cores = 1,
  return.full.beta = FALSE,
  silent = TRUE,
  fold_assign = NULL
)

Arguments

x	Design matrix, n x p.
y	Vector of responses, length n.
K	Number of folds for cross-validation. Must be at least 2
var_order	For user-specified ordering of variables. Indices start at 0, start with least important variable and end with most. By default order will be induced from scaling of columns in design matrix
lambda	For user-specified sequence of tuning parameter lambda
nlambda	Length of automatically generated sequence of tuning parameters lambda
grid.size	Number of subsets of variables for which a solution path will be computed for
lambda.min.ratio	Ratio of max/min lambda for automatically generated sequence of tuning parameters lambda
thresh	Convergence threshold for coordinate descent for difference in objective values between successive iterations
maxit	Maximum number of iterations for coordinate descent routine
mc.cores	Number of cores to be made available for computing the cross-validation estimates in parallel
return.full.beta	Return the entire solution path for the chosen variable subset, as opposed to only the estimate for estimated optimal lambda
silent	Suppress some text to console
fold_assign	For user-specified vector of assignment of folds for cross-validation. Must be of the form of integer vector with entries in 1 , ... , K.

Value

A glmnet model object, with some additional attributes:

• best – an index denoting which point on the path of lambda values is estimated optimal

• cv – matrix of cross-validation error for each value of lambda and gridpoint

Examples

set.seed(1)
X = matrix(0, 50, 500)
betavec = c(rep(1,5),rep(0,495))
X[ , 1:5 ] = matrix(rnorm(250), 50, 5)
Y = as.vector(X %*% betavec)
Y = Y + rnorm(50)
Y = as.numeric(Y >= mean(Y))
X = X + matrix(rnorm(50*500), 50, 500)
mod1 = cv_grid_lasso_logistic(X, Y, grid.size = 25)

bgs25/SubsetGridRegression documentation built on Dec. 19, 2021, 8:50 a.m.