Biometrical Letters vol. 45(2), 2008, pp. 43-55
|
NOTES ON THE USE OF THE GOLDFELD-QUANDT TEST FOR HETEROSCEDASTICITY IN ENVIRONMENT RESEARCH A. Budka, D. Kachlicka, M. Kozłowska Department of Mathematical and Statistical Methods, Poznan University of Life Sciences, Wojska Polskiego 28, 60 637 Poznań, Poland, e-mail: budka@up.poznan.pl |
Some methods of using the Goldfeld-Quandt test are described. The use of classical statistics (mean, standard deviation) or position statistics (median, average deviation, median absolute deviation) is proposed. The results can be used to test the hypothesis that the residuals from a linear regression are homoscedastic, when the experimental distribution of the considered variable is symmetrical or asymmetrical (right-sided skew or left-sided skew). Using Monte Carlo method, properties of these modifications of the Goldfeld-Quandt procedure are explored. A comparison of the effectiveness of the described methods for environment research is presented.
Goldfeld-Quandt test, heteroscedasticity, linear regression, position statistics
