Description
Cross-Validation to estimate regularization parameters for
sparse inverse covariance estimation.Usage
CVcov(x, maxlam, minlam, steps, pmiss = 0.01, do = 2, trace = TRUE)
Arguments
maxlam
Maximum regularization parameter.
minlam
Minimum regularization parameter.
steps
Number of regularization parameters to test.
pmiss
Percentage missing in each fold.
do
Number of folds. Note that for medium or large size data
matrices, often one fold is sufficient.
trace
Logical. Output the penalized log-likelihood and MSE for
each step and fold.
Value
- cvmatMatrix of cross-validation mean squared errors.
- optlamOptimal value of the regularization parameter as
estimated by cross-validation.
- lamsValues of the regularization parameters tested.
References
G. I. Allen and R. Tibshirani, "Transposable regularized covariance
models with an application to missing data imputation", Annals of
Applied Statistics, 4:2, 764-790, 2010.