Comments on: Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination

Abstract

The authors are to be commended for bringing the critical problem of cellwise outliers to the attention of a broader community and providing some important new estimation methods and related theory. High dimensional data analysis has become a critical area for research in statistical theory and practice and, in such situations, removing (or severely downweighting) an entire observation for a single cellwise outlier can eliminate most of the data.