Partial Correlation Graphical LASSO

Abstract

Standard likelihood penalties to learn Gaussian graphical models are based on regularising the off-diagonal entries of the precision matrix. Such methods, and their Bayesian counterparts, are not invariant to scalar multiplication of the variables, unless one standardises the observed data to unit sample variances. We show that such standardisation can have a strong effect on inference and introduce a new family of penalties based on partial correlations. We show that the latter, as well as the maximum likelihood, $L_0$ and logarithmic penalties are scale invariant. We illustrate the use of one such penalty, the partial correlation graphical LASSO, which sets an $L_1$ penalty on partial correlations. The associated optimization problem is no longer convex, but is conditionally convex. We show via simulated examples and in two real datasets that, besides being scale invariant, there can be important gains in terms of inference.

41 pages, 7 figures

References in corpus (8)

• Nearly unbiased variable selection under minimax concave penalty
• High-dimensional graphs and variable selection with the Lasso
• Extended Bayesian Information Criteria for Gaussian Graphical Models
• One-step sparse estimates in nonconcave penalized likelihood models
• Artificial Intelligence and Statistics
• Network exploration via the adaptive LASSO and SCAD penalties
• Adaptive Lasso for High Dimensional Regression and Gaussian Graphical Modeling
• Discussion: One-step sparse estimates in nonconcave penalized likelihood models