ROBUST M-ESTIMATION OF A DISPERSION MATRIX WITH A STRUCTURE

ROBUST M-ESTIMATION OF A DISPERSION MATRIX
WITH A STRUCTURE

MADHUSUDAN BHANDARY

Department of Statistics, North Dakota State University, Fargo, ND 58105, U.S.A.

(Received June 2, 1988; revised April 13, 1990)

Abstract. An iterative algorithm for the robust M-estimation of the dispersion matrix of the form Gamma + sigma²I_p has been given. This algorithm converges after some steps and reduces the effect of outliers on the covariance matrix. The consistency and asymptotic normality of the estimator are established.

Key words and phrases: Robust estimation, M-estimation, elliptically symmetric distribution, outliers, signal processing.

Source ( TeX , DVI , PS )