A Procedure for Detecting a Pair of Outliers in Multivariate Dataset

B. K. Nkansah, B. K. Gordor

Abstract


The paper presents a procedure for detecting a pair of outliers in multivariate data. The procedure involves a reduction of the dimensionality of the dataset to only two dimensions along outlier displaying components, and then determines the orientation of a least squares ellipse that fits the scatter of points of the two dimensional dataset. Finally, the reduced data is projected unto a vector which is determined in terms of the orientation of the ellipse. The results show that if two observations constitute a pair of outliers in a data set, then the pair is extreme at either ends of the one-dimensional projection and separated clearly from the remaining observations. If the two outliers are not distinct on such a one-dimensional projection, three key rules are prescribed for successful determination of the right pair of outliers.

Key words: Multiple Outlier Detection; Outlier Displaying Component


Keywords


Multiple Outlier Detection; Outlier Displaying Component

Full Text:

PDF


DOI: http://dx.doi.org/10.3968/j.sms.1923845220120402.1623

DOI (PDF): http://dx.doi.org/10.3968/g2562

Refbacks

  • There are currently no refbacks.


Copyright (c)




Share us to:   


Please send your manuscripts to sms@cscanada.net,or  sms@cscanada.org  for consideration. We look forward to receiving your work.

 

 Articles published in Studies in Mathematical Sciences are licensed under Creative Commons Attribution 4.0 (CC-BY).

 STUDIES IN MATHEMATICAL SCIENCES Editorial Office

Address: 1055 Rue Lucien-L'Allier, Unit #772, Montreal, QC H3G 3C4, Canada.

Telephone: 1-514-558 6138

Http://www.cscanada.net
Http://www.cscanada.org
E-mail:caooc@hotmail.com

Copyright © 2010 Canadian Research & Development Centre of Sciences and Cultures