Sparse permutation invariant covariance estimation

A method for constructing a sparse estimator for the inverse covariance (concentration) matrix in high-dimensional settings using a penalized normal likelihood approach and forces sparsity by using a lasso-type penalty is proposed.