A COMPARATIVE PERFORMANCE OF SEVERAL ROBOST DISCRIMINANT ANALYSIS PROCEDURES

Title of Publication: A COMPARATIVE PERFORMANCE OF SEVERAL ROBOST DISCRIMINANT ANALYSIS PROCEDURES
Author(s): Okonkwo E.N, Onyeagu S.I, Okeke J.U
Year of Publication: 2011
Abstract
A nonparametric discriminant analysis is discussed and used to compare two robust linear discriminant functions that are based on high breakdown point (of location and covariance matrix ) estimators. The major part of the paper deals with practical application of the available methods. The methods are compared on 10 simulated data sets. We concluded that the nonparametric procedure have the highest predictive power among other procedures we considered.

Key words: discriminant analysis (DA), MCD, MWCD, principal component analysis,projection pursuit, Principal component and Projection Pursuit(Nonparametric DA)

Need complete download?
Login to download
NSA Executives

President :: Prof. S.I. Onyeagu
1st Vice President :: Dr. G.U. Ebuh
2nd Vice President :: Dr. O.E. Olubusoye
National Secretary :: Mr. O.T. Arowolo
Assistant National Secretary :: Mr. C. O. Aronu
National Financial Secretary :: Mr. R.O. Olaitan
National Treasurer :: Mrs. F.O. Obikudu
Business Manager :: Mr. B.F. Ajibade
Editor-in-chief :: Prof. J.I. Mbegbu
Ordinary Council Member 1 :: Elder E. E. Umanah
Ordinary Council Member 2 :: Prof. S. U. Gulumbe
Immediate Past President :: Dr. M.M. Tumala
Editorial Team

Admin

Feed Back
We will love to hear from you

Planners, Demographers, Analysts and Corporate Organisations with interest in Statistics.

You must be logged in to use our feedback system



Publish with Us

The Nigerian Statistical Association was founded in 1976 as a non-profit making professional body of Statisticians, Economists.

Feel free to drop us a note
Unsplashed background img 3
Send us a message