Pathwise Calibrated Sparse Shooting Algorithm
Description
Implements the pathwise calibrated sparse shooting algorithm (PICASSO) for high dimensional regularized sparse linear regression, sparse logistic regression, and sparse undirected graphical model estimation. As a pathwise coordinate optimization algorithm, PICASSO exploits warm start initialization, active set updating, and strong rule for coordinate preselection to boost computation, and attains a linear convergence to a unique sparse local optimum with optimal statistical properties. Moreover, users can choose different regularizers, including L1, MCP, and SCAD regularizers. The computation is memory-optimized using the sparse matrix output.