The R-package asus implements the ASUS (Adpative SURE thresholding with Side Information) procedure that estimates a high-dimensional sparse parameter when along with the primary data we can also gather side information from secondary data sources. ASUS is an adaptive and robust methodology for leveraging auxiliary data to improve the accuracy... [Read More]

The goal of fusionclust is to conduct clustering and feature screening in large scale cluster analysis problems. In particular, fusionclust provides the Big Merge Tracker (BMT) and COSCI algorithms for convex clustering and feature screening using an ℓ1 fusion penalty. [Read More]

COSCI (COnvex Screening for Cluster Information) is a non-parametric method for ranking and screening non-informative features in large scale cluster analysis problems. Unlike the non-parametric density estimation based screening techniques, COSCI is very scalable, and can successfully handle datasets with more than one million observations. COSCI discards non-informative features by... [Read More]

BMT (Big Merge Tracker) minimizes the sample within cluster sum of squares under an $\ell_1$ fusion constraint on the cluster centroids. The codes for implementing BMT as well as the codes for producing the numerical results in the paper can be downloaded from here. [Read More]

R codes for the paper “Exact minimax estimation of the predictive density in sparse Gaussian models”. Numerical evaluations of the risk functions depicted in Figure 3 of the paper and for calculating the cardinality of positive support points of the cluster prior reported in Table 1 of the paper, can... [Read More]