B0857
Title: Robust and resistant regularized covariance matrices
Authors: David Tyler - Rutgers (United States)
Mengxi Yi - Bejing Normal Univeristy (China)
Klaus Nordhausen - University of Helsinki (Finland) [presenting]
Abstract: A class of regularized M-estimators of multivariate scatter is introduced. The scatter matrices possess high breakdown points analogous to the popular spatial sign covariance matrix (SSCM). It is also shown that the SSCM can be viewed as an extreme member of this class. Unlike the SSCM, this class of estimators takes into account the shape of the contours of the data cloud when down-weighing observations. In addition, a median-based cross-validation criterion is proposed for selecting the tuning parameter for this class of regularized M-estimators. This cross-validation criterion helps ensure the resulting tuned scatter estimator is a good fit to the data, as well as having a high breakdown point.
