capr - Covariate Assisted Principal Regression
Covariate Assisted Principal Regression (CAPR) for multiple covariance-matrix outcomes. The method identifies (principal) projection directions that maximize the log-likelihood of a log-linear regression model of the covariates. See Zhao et al. (2021), "Covariate Assisted Principal Regression for Covariance Matrix Outcomes" <doi:10.1093/biostatistics/kxz057>.
Last updated
openblascpp
4.82 score 102 scripts 145 downloadsclime - Constrained L1-Minimization for Inverse (Covariance) Matrix Estimation
A robust constrained L1 minimization method for estimating a large sparse inverse covariance matrix (aka precision matrix), and recovering its support for building graphical models. The computation uses linear programming. The method was published in TT Cai, W Liu, X Luo (2011) <doi:10.1198/jasa.2011.tm10155>.
Last updated
4.16 score 3 dependents 32 scripts 361 downloadscord - Community Estimation in G-Models via CORD
Partitions data points (variables) into communities/clusters, similar to clustering algorithms such as k-means and hierarchical clustering. This package implements a clustering algorithm based on a new metric CORD, defined for high-dimensional parametric or semiparametric distributions. For more details see Bunea et al. (2020), Annals of Statistics <doi:10.1214/18-AOS1794>.
Last updated
cpp
2.00 score 3 scripts 247 downloadslorec - LOw Rand and sparsE Covariance matrix estimation
Estimate covariance matrices that contain low rank and sparse components
Last updated
openblasfortran
1.00 score 1 scripts 183 downloads