Proceedings of the International Academy of Ecology and Environmental Sciences, 2018, 8(1): 78-82
[XML] [EndNote] [RefManager] [BibTex] [ Full PDF (73K)] [Comment/Review Article]


One-way ANOVA and comfortless questions: Direct computer experiment

L.V. Nedorezov
Research Center for Interdisciplinary Environmental Cooperation RAS, nab. Kutuzova 14, Saint-Petersburg, 191187, Russia

Received 20 November 2017;Accepted 20 December 2017;Published 1 March 2018

For matrix 5x5 (5 gradations of a factor with 5 values of characteristics in every gradation) values of normally distributed stochastic variables were generated with equal averages (for all gradations of a factor) and equal variances. Results of application of Kolmogorov's test to samples of variance ratios (for checking of correspondence of samples to Fisher distribution) are presented in table for 5% significance level. Every element of table was obtained after analysis of 107 independent samples. Changing of elements of table was analyzed with Theil criterion. Obtained results demonstrate that there exist some problems with identification of distribution of variance ratio and in analysis of its correspondence to Fisher distribution.

Keywords one-way ANOVA;Kolmogorov's test;distribution of variance ratio;Theil criterion.

International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2020 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: website@iaees.org; Last modified: 2020-9-24

Translate page to: