SparseBiplots - 'HJ-Biplot' using Different Ways of Penalization Plotting with 'ggplot2'.
'HJ-Biplot' is a multivariate method that allow represent multivariate data on a subspace of low dimension, in such a way that most of the variability of the information is captured in a few dimensions. This package implements three new techniques and constructs in each case the 'HJ-Biplot', adapting restrictions to reduce weights and / or produce zero weights in the dimensions, based on the regularization theories. It implements three methods of regularization: Ridge, LASSO and Elastic Net.
Last updated 3 years ago
3.00 score 2 stars 1 scripts 138 downloads