rsvddpd: Robust Singular Value Decomposition using Density Power Divergence

Computing singular value decomposition with robustness is a challenging task. This package provides an implementation of computing robust SVD using density power divergence (<arXiv:2109.10680>). It combines the idea of robustness and efficiency in estimation based on a tuning parameter. It also provides utility functions to simulate various scenarios to compare performances of different algorithms.

install.packages("rsvddpd")

Try the rsvddpd package in your browser