Little, Roderick J. A. Robust estimation of the mean and covariance matrix from data with missing values. (English) Zbl 0647.62040 J. R. Stat. Soc., Ser. C 37, No. 1, 23-38 (1988). Summary: Methods of D. B. Rubin [Iteratively reweighted least squares. Entry in S. Kotz, N. L. Johnson and C. B. Read (eds.), Encyclopedia of statistical sciences, Vol. 4 (1983; Zbl 0585.62002)] for robust estimation of a mean and covariance matrix and associated parameters are extended to analyse data with missing values. The methods are maximum likelihood (ML) for multivariate t and contaminated normal models. ML estimation is achieved by the EM algorithm, and involves minor modifications to the EM algorithm for multivariate normal data. The methods are shown to be superior to existing methods in a simulation study, using data generated from a variety of models. Model selection and standard error estimation are discussed with the aid of two real data examples. Cited in 49 Documents MSC: 62F35 Robustness and adaptive procedures (parametric inference) 62H12 Estimation in multivariate analysis Keywords:multivariate linear regression; mixture models; robust estimation; mean; covariance matrix; missing values; maximum likelihood; multivariate t; contaminated normal models; EM algorithm; multivariate normal data; simulation study; Model selection; error estimation Citations:Zbl 0585.62002 PDFBibTeX XMLCite \textit{R. J. A. Little}, J. R. Stat. Soc., Ser. C 37, No. 1, 23--38 (1988; Zbl 0647.62040) Full Text: DOI