gcrit() calculates the critical value of the GGL objective funciton. It is used to confirm that the GGL algorithm is converging.

1 | ```
gcrit(theta, S, n, lam1, lam2, penalize.diagonal)
``` |

`theta` |
A list of pXp inverse covariance matrices. |

`S` |
A list of pXp empirical covariance matrices. |

`n` |
A vector of sample sizes to attribute to each of the K data matrices. n controls the relative weights of the classes: for example, with n==c(1,1), each class's theta will be penalized equally. |

`lam1` |
The tuning parameter for the graphical lasso penalty. |

`lam2` |
The tuning parameter for the group lasso penalty. |

`penalize.diagonal` |
Logical, determining whether the penalties will be applied to the diagonal elements of the theta matrices. |

A function called by GGL to calculate the critical value of the objective function.

crit, the critical value of the list of inverse covariance matrices.

Patrick Danaher

Patrick Danaher, Pei Wang and Daniela Witten (2011). The joint graphical lasso for inverse covariance estimation across multiple classes. http://arxiv.org/abs/1111.0324

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.