Asymptotic theory for robust principal components (Q1088340)

The asymptotic distribution of the eigenvalues and eigenvectors of the robust scatter matrix, proposed by \textit{R. A. Maronna} [Ann. Stat. 4, 51- 67 (1976; Zbl 0322.62054)], is given when the observations are from an elliptical distribution. The elements of each characteristic vector are the coefficients of a robustified version of principal components. A definition for the asymptotic efficiency of the estimators is given and their influence curve is evaluated. The problem of maximizing the efficiency under a bound on the influence curve is solved.