cv.hmlasso: Fit a model using a design matrix with cross validation

In hmlasso: Lasso with High Missing Rate

Description

Fit a model using a design matrix with cross validation

Usage

cv.hmlasso(X, y, nfolds = 10, lambda.min.ratio = 0.01, nlambda = 100,
  lambda = NULL, foldid = NULL, unit = "sample", seed = 0,
  min_eig_th = 1e-06, use = "pairwise.complete.obs",
  impute_method = "mean", direct_prediction = FALSE,
  adjust_by_tr = FALSE, positify = "diag", weight_power = 1,
  mu = 1, eig_tol = 1e-08, eig_maxitr = 1e+08, verbose = FALSE,
  ...)

Arguments

X	matrix of explanatory variables
y	vector of objective variable
nfolds	the number of folds (ignored if foldid is specified)
lambda.min.ratio	ratio of max lambda and min lambda (ignored if lambda is specified)
nlambda	the number of lambda (ignored if lambda is specified)
lambda	lambda sequence
foldid	vector indicating id of fold for each sample
unit	unit for cross validation error: "sample" (default) or "fold"
seed	random seed of cross validation
min_eig_th	minimum eigenvalue
use	method to calculate correlation matrix from missing data (default "pairwise.complete.obs")
impute_method	imputation method for predictions
direct_prediction	either corrected cross validation is used or not
adjust_by_tr	whether mean (or median) of training data for prediction is used or not
positify	method for solving PSD matrix
weight_power	weighting power (default 0 meaning no-weighting)
mu	augmented Lagrangian parameter
eig_tol	tol parameter in eigs_sym function
eig_maxitr	maxitr parameter in eigs_sym
verbose	whether output verbose warnings and messages (default FALSE)
...	parameters of hmlasso function

Value

lasso model

fit	lasso model with hole data
lambda.min	lambda with minimum cross validation error
lambda.min.index	index of lambda.min
lambda.1se	largest lambda such that error is within 1 standard error of the minimum
lambda.1se.index	index of lambda.1se
foldid	fold id
cve	cross validation error
cvse	cross validation standard error
cvup	cross validation error + standard error
cvlo	cross validation error - standard error
pe	prediction error (for family="binomial")

Examples

X_incompl <- as.matrix(iris[, 1:3])
X_incompl[1:5,1] <- NA
X_incompl[6:10,2] <- NA
y <- iris[, 4]
cv_fit <- cv.hmlasso(X_incompl, y, nlambda=50, lambda.min.ratio=1e-2)
plot(cv_fit)
plot(cv_fit$fit)

hmlasso documentation built on Aug. 3, 2019, 9:03 a.m.