Accession Number : ADA130686

Title :   Rejection of Multivariate Outliers.

Descriptive Note : Technical rept.,

Corporate Author : PITTSBURGH UNIV PA CENTER FOR MULTIVARIATE ANALYSIS

Personal Author(s) : Sinha,Bimal Kumar

PDF Url : ADA130686

Report Date : May 1983

Pagination or Media Count : 31

Abstract : An extension of Ferguson's univariate normal results for rejection of outliers is made to the multivariate case with mean slippage. The formulation is more general than that in Schwager and Margolin and the approach is also different. The main result can be viewed as a robustness property of Mardia's locally optimum multivariate normal kurtosis test to detect outliers against nonnormal multivariate distributions.

Descriptors :   *Multivariate analysis, Normal distribution, Invariance, Transformations(Mathematics), Matrices(Mathematics), Probability distribution functions, Statistical analysis, Modification, Statistical inference, Computations, Rejection, Mathematical models

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE