Linear discriminant function under multivariate non-normal rice (Oryza sativa) and maize (Zea mays) data
194 / 66
Authors
-
R K RAMAN
Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi 110 012
-
S D WAHI
Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi 110 012
-
A K PAUL
Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi 110 012
Keywords:
Discriminant analysis, Multivariate non-normal distribution, Probability of misclassificationAbstract
The performance of linear discriminant function was studied under multivariate non-normal situations. The different multivariate non-normal populations were simulated by using the mean vectors and dispersion matrices of rice (Oryza sativa L.) and maize (Zea mays L.) data sets. Further 50 different independent samples were simulated for different dimensions and sample sizes for maize and rice data to obtain empirical probabilities of misclassification. On fitting linear discriminant function to non-normal data the empirical probabilities of misclassification were higher as compared to misclassifying probabilities obtained by using normal approximation. In large sample sizes and in higher dimensions the differences between empirical and normal approximation of probabilities of misclassification were found almost negligible.
Downloads
References
Fisher R A. 1936. The use of multiple measurements in taxonomic problems. Ann. Eugenics 7: 179–88.
Ito K and Schull W J. 1964. On robustness of T 2 in multivariate analysis of variance when covariance matrices are not equal, Biometrica 51: 71–82.
Kotz S, Balakrishnan N and Jhonson N L. 2000. Continuous Multivariate Distributions, pp 485–41. John Wiley & Sons, Inc., New York.
Mardia K V. (1980). (in) Handbook of Statistics, 1: 279–320, P R Krishnaiah (Ed.) Test of univariate and multivariate normality. North Holland.
Minhajuddin A M, Harris I R and Schucany W R. 2004. Simulating multivariate distribution with specific correlation. Journal of Statistical Computation and Simulation 74(8): 599– 607.
Rausch R J and Kelley. 2009. A comparison of linear and nonlinear models for discriminant analysis under non-normality. Behaviour
Research Methods 41(1): 85–98.
SAS. 2009. SAS User Guide, Version 9.2, SAS Institute Incorporates, USA.
Sever M, Lajovic J and Rajer B. 2005. Robustness of Fisher’s discriminant function to skewed normal distribution. Metodoliski Zvezki 2(2): 231–42.
Downloads
Submitted
Published
Issue
Section
License
Copyright (c) 2014 The Indian Journal of Agricultural Sciences
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
The copyright of the articles published in The Indian Journal of Agricultural Sciences is vested with the Indian Council of Agricultural Research, which reserves the right to enter into any agreement with any organization in India or abroad, for reprography, photocopying, storage and dissemination of information. The Council has no objection to using the material, provided the information is not being utilized for commercial purposes and wherever the information is being used, proper credit is given to ICAR.
How to Cite
