Code

TVVAR: regularized estimation of sparse transition matrices of the time-varying vector autoregressive model. R implementation of the paper.

MMT: Matlab library for Multi-Task learning. Implementation of the multi-task Lasso models, including: (i) the Lasso; (ii) the standard multi-task Lasso, i.e. the group Lasso; (iii) the structured input-output multi-task Lasso, a.k.a. the two-graph guided multi-task Lasso.

Least Angle Regression (LARS): Matlab code for the LARS algorithm, which computes the whole optimal path, by a homotopy approach, for the LAR and Lasso problem in constrained form.

Lasso and group Lasso: Shooting algorithm implemented in Matlab for solving the Lasso and group Lasso problems in the penalized form.

Graphical Lasso: Matlab implementation of the graphical Lasso model for estimating sparse inverse covariance matrix (a.k.a. precision or concentration matrix).