Empirical Type 1 Error Rate and Power Comparisons of Normality Tests with R

Ahmet Pekgör; Murat Erişoğlu; Aydın Karakoca; Ülkü Erişoğlu

doi:10.17776/csj.426382

EN TR

Empirical Type 1 Error Rate and Power Comparisons of Normality Tests with R

Abstract

Normality is one of the main presuppositions in statistical tests. The multiplicity of the normality tests bring out another problem of choosing the appropriate test for researchers. The free software R which has a great popularity in the statistical analysis has 18 normality tests in 4 different packages. In this study we compared performance of these normality tests in terms of empirical type 1 error rate and power by Monte Carlo simulation. As a result, regardless of the distribution of data (symetric or asymmetric) the Shapiro-Francia test, also the Frosini B test performed better than the other normality tests in terms of experimental type 1 error rate. However the widely used Kolmogorov-Smirnov test showed worse performance than other normality tests in terms of empirical type 1 error rate and power.

Keywords

Normality tests,R,Empirical type 1 error rate,Power of test

References

[1] Pearson, K., X. on the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 50-302 (1900) 157–175.
[2] Von Mises, R., Wahrscheinlichkeitsrechnung und ihre Anwendung in der Statistik und theoretischen Physik. Bull. Amer. Math. Soc 38 (1932) 169-170.
[3] Kolmogorov, A., Sulla determinazione empirica di una lgge di distribuzione, Inst. Ital. Attuari, Giorn. 4 (1933) 83–91.
[4] Geary, R. C., The ratio of the mean deviation to the standard deviation as a test of normality, Biometrika 27-3/4 (1935) 310–332.
[5] Anderson, T. W. and Darling, D. A., A test of goodness of fit, Journal of the American statistical association 49-268 (1954) 765–769.
[6] Kuiper, N. H., Tests concerning random points on a circle, Nederl. Akad. Wetensch. Proc. Ser. A, Vol. 63, (1960) 38–47.
[7] Shapiro, S. S. and Wilk, M. B., An analysis of variance test for normality (complete samples), Biometrika 52-3/4 (1965) 591–611.
[8] Lilliefors, H. W., On the kolmogorov-smirnov test for normality with mean and variance unknown, Journal of the American statistical Association 62-318 (1967) 399–402.

[9] Shapiro, S. S. and Francia, R., An approximate analysis of variance test for normality, Journal of the American Statistical Association 67-337 (1972) 215–216.
[10] Hegazy, Y. and Green, J., Some new goodness-of-fit tests using order statistics, Applied Statistics pp. (1975) 299–308.
[11] Weisberg, S. and Bingham, C., An approximate analysis of variance test for non-normality suitable for machine calculation, Technometrics 17-1 (1975) 133–134.
[12] Spiegelhalter, D., A test for normality against symmetric alternatives, Biometrika 64-2 (1977) 415–418.
[13] Frosini, B. V., A survey of a class of goodness-of-fit statistics, Universit`a degli studi. Facolt`a di scienze statistiche, demografiche ed attuariali (1978).
[14] Jarque, C. M. and Bera, A. K., A test for normality of observations and regression residuals, International Statistical Review/Revue Internationale de Statistique (1987) 163–172.
[15] Urzua, C. M., On the correct use of omnibus tests for normality, Economics Letters 3-54 (1997) 301.
[16] Esteban, M., Castellanos, M., Morales, D. and Vajda, I., Monte carlo comparison of four normality tests using different entropy estimates, Communications in Statistics-Simulation and computation 30-4 (2001) 761–785.
[17] Yazici, B. and Yolacan, S., A comparison of various tests of normality, Journal of Statistical Computation and Simulation 77-2 (2007) 175–183.
[18] Yap, B. W. and Sim, C. H.,Comparisons of various types of normality tests, Journal of Statistical Computation and Simulation 81-12 (2011) 2141–2155.
[19] Noughabi, H. A. and Arghami, N. R. ,Monte carlo comparison of seven normality tests, Journal of Statistical Computation and Simulation 81-8 (2011) 965–972.
[20] Royston, J., An extension of shapiro and wilk’s w test for normality to large samples, Applied statistics pp. (1982) 115–124.
[21] Razali, N. M., & Wah, Y. B.. Power comparisons of shapiro-wilk, kolmogorov-smirnov, lilliefors and anderson-darling tests. Journal of statistical modeling and analytics, 2-1(2011) 21-33.

Details

Primary Language

English

Subjects

-

Journal Section

Research Article

Authors

Ahmet Pekgör
Türkiye

Murat Erişoğlu
Türkiye

Aydın Karakoca
0000-0001-6503-3872
Türkiye

Ülkü Erişoğlu
Türkiye

Publication Date

September 30, 2018

Submission Date

May 23, 2018

Acceptance Date

September 3, 2018

Published in Issue

Year 1970 Volume: 39 Number: 3

DOI

https://doi.org/10.17776/csj.426382

IZ

https://izlik.org/JA77AH27BH

Cite

RIS / Bibtex

APA

Pekgör, A., Erişoğlu, M., Karakoca, A., & Erişoğlu, Ü. (2018). Empirical Type 1 Error Rate and Power Comparisons of Normality Tests with R. Cumhuriyet Science Journal, 39(3), 799-811. https://doi.org/10.17776/csj.426382

Empirical Type 1 Error Rate and Power Comparisons of Normality Tests with R

Empirical Type 1 Error Rate and Power Comparisons of Normality Tests with R

Abstract

Keywords

R ile Normallik Testlerinin Deneysel 1. Tip Hata Oranı ve Güç Karşılaştırması

Öz

Anahtar Kelimeler

References

Details

Primary Language

Subjects

Journal Section

Authors

Publication Date

Submission Date

Acceptance Date

Published in Issue

DOI

IZ

Cite