Accession Number : ADA119659

Title :   Self-Critical, and Robust, Estimates for the Parameters of the Multivariate Normal Distribution.

Descriptive Note : Interim technical rept.,

Corporate Author : RENSSELAER POLYTECHNIC INST TROY NY

Personal Author(s) : Delaney,N J ; Paulson,A S

PDF Url : ADA119659

Report Date : 01 Jun 1982

Pagination or Media Count : 16

Abstract : This algorithm yields joint robust estimates of the location vector and the variance-covariance matrix for samples from the multivariate normal distribution. The degree of robustness depends on a single filtering parameter, c, set by the user. The algorithm provides, for each observation, an internally determined weight which may be used to identify potential outliers.

Descriptors :   *Multivariate analysis, *Normal distribution, *Estimates, Algorithms, Parameters, Analysis of variance, Covariance, Matrices(Mathematics), Vector analysis, Position(Location), Weighting functions, Observation

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE