A Brief Study about Nonparametric Adherence Tests

The statistical study has become indispensable for various fields of knowledge. Not any different, in Geotechnics the study of probabilistic and statistical methods has gained power considering its use in characterizing the uncertainties inherent in soil properties. One of the situations where engineers are constantly faced is the definition of a probability distribution that represents significantly the sampled data. To be able to discard bad distributions, goodness-of-fit tests are necessary. In this paper, three non-parametric goodness-of-fit tests are applied to a data set computationally generated to test the goodness-of-fit of them to a series of known distributions. It is shown that the use of normal distribution does not always provide satisfactory results regarding physical and behavioral representation of the modeled parameters.

Authors:

• Vinicius R. Domingues
• Luan C. S. M. Ozelim

Keywords:

• Kolmogorov-Smirnov
• Anderson-Darling
• Cramer-Von-Mises
• Nonparametric adherence tests.

References:

[1] Torman, V. B. L.; Coster, R.; Riboldi, J. 2012. Normalidade de
variáveis: métodos de verifcação e comparação de alguns testes não
paramétricos por simulação, Rev. HCPA, Porto Alegre, v.32, n.2, p.227-
234.
[2] Anderson, T. W. and Darling, D. A. 1954. A Test of Goodness-of-Fit.
Journal of the American Statistical Association, 49: 765–769.
[3] Gassem, A. 2011. On Cramér–von Mises type test based on local time of
switching diffusion process. Journal of Statistical Planning and
Inference, Vol 141(4), P. 1355–1361.
[4] Inglot, T. and Ledwina, T. 2004. On consistent minimax
distinguishability and intermediate efficiency of Cramér–von Mises test.
Journal of Statistical Planning and Inference, Vol. 124 (2), P. 453–474.
[5] D'Agostino, R. B. and Stephens, M. A.. 1986. Goodness-of-Fit
Techniques. New York: Marcel Dekker. ISBN 0-8247-7487-6, 1986.
[6] Coronel-Brizio, H. F. and Hernández-Montoya, A. R. 2010. The
Anderson–Darling test of fit for the power-law distribution from leftcensored
samples. Physica A: Statistical Mechanics and its Applications,
Vol. 389 (17), P. 3508–3515.
[7] Heo, J-H., Shin, H., Nam, W., Om, J., Jeong, C. 2013. Approximation of
modified Anderson–Darling test statistics for extreme value distributions
with unknown shape parameter. Journal of Hydrology, Vol. 499 (30), P.
41–49.