Robust Estimation under Cellwise and Casewise Contamination

In traditional robust statistics, it is generally assumed that the majority of the observations in the data are free of contamination, while only a minority of the observations are contaminated. The contaminated observations are flagged as outliers and down-weighted even if only a single component is contaminated. Some observations may fully depart from the bulk of the data. This situation usually refers to casewise outliers. However, observations can be only partially contaminated. This type of contamination often appears as single outlying cells in a data matrix and therefore, usually refers to cellwise contamination. Under cellwise contamination, a lot of information could be lost through down-weighting the whole observation, especially for high-dimensional data. Recent work has shown that procedures that proceed in such way are not robust. In this talk, we will sketch out our proposal to estimate multivariate location and scatter under this cell-and-casewise contamination.

Date and Time: 

Wednesday, June 14, 2017 - 14:15 to 15:00

Co-authors (not including you): 

Victor J. Yohai
Universidad de Buenos Aires
Ruben H. Zamar
University of British Columbia

Language of Oral Presentation: 


Language of Visual Aids: 


Type of Presentation: 




First Name Middle Name Last Name Primary Affiliation
Andy Leung The University of British Columbia