Resumo

MARCANO, Luis  e  FERMIN, Wilmer. COMPARISON OF MULTIVARIATE METHODS FOR OUTLIERS DETECTION BY SIMULATION. Saber [online]. 2013, vol.25, n.2, pp.193-201. ISSN 1315-0162.

Outliers constitute a constant problem in data collection, they are observations that deviate from the general pattern of the rest of the data and thus can affect the results that derive from the application of univariate and multivariate statistical methods. It is essential to detect these observations, either to eliminate them or to mitigate their  effect  on  the  analysis.  Several  outlier  detection  methods  have  been  developed,  including  the  Robust Mahalanobis Distance (DRB) by Rousseeuw and Van Zomeren (1990), the Kurtosis-1 by Peña and Prieto (2001) and  the  FGR  method  by  Filzmoser,  Garrett  y  Reimann  (2005).  These  three  methods  were  compared  in  this article, in five correlation scenarios considering explanatory variables with several percentages of outliers, by using comparative analysis of these methods in simulated data. Results show that the kurtosis-1 method is more efficient than DRM and FGR for the detection of multivariate outliers, regardless the proportion of outliers and the presence of correlation among variables in the research study.

Palavras-chave : Multivariate outliers; detection; comparison; simulation.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )