The  eenvlp  is an R package for multivariate regression that provides an enhanced response envelope estimator for the regression coefficients. This estimator improves prediction accuracy by capturing the core lower-dimensional structure within the responses. It can be used when the number of responses is smaller than the sample size, regardless of the number of predictors. [website] [paper]

The  euspca  is an R package for finding exactly uncorrelated sparse principal components. [website] [paper]

The  lhsc  is an R package for binary classification problems. It provides a classifier derived by the leaky hockey stick loss function, which penalizes incorrect predictions and rewards correct ones. This is different from existing loss functions that only penalize mistakes. [website] [paper]