Comparison between some parametric robust methods for estimating the parameters of the multiple normal distribution

الملخص EN

Estimating the parameters of the Multivariate Normal Distribution is very important process in many statistical Application like the Principal Component Analysis or Canonical Analysis.

The paper aims at finding robust and efficient estimators for the parameters of the multivariate normal distribution by using parametric method which is the reweighted minimum vector method RMV and compare it with another robust estimation methods like S estimation method in case of deferent sample sizes and deferent contaminated ratios.

Results shows that the reweighted minimum vector is the best method via simulation and real data was taken from waist sewerage directorate minimum esquire error (MMSE) was used as a comparison tool between the two estimation methods.