A test for homogeneity of variances using the jackknife empirical likelihood approach and its comparison with several common tests

Authors
Nabaz Esmailzadeh
Parisa Hosini
University of Kurdistan
Abstract
The homogeneity of variances test is a prerequisite for many statistical methods. In this article, a recently introduced test based on the jackknife approach is compared with common tests such as Levene's and Bartlett's tests, as well as two tests by James and Alexander-Govern, in terms of their ability to maintain the first type error rate and test power for several distributions. The permutation versions of these tests were also examined. The results indicate that the performance of the tests significantly improves in the permutation version. To evaluate the performance of the tests in the real world, the tests were applied to two real data sets, and the results are presented.
Keywords
Homogeneity of variances
Levene's test
Jackknife
Permutation test
Bartlett's tests
[1] Alexander, R.A. and Govern, D.M. (1994). A new and simpler approximation and ANOVA under variance heterogeneity. J. Edue. Stat. 19, 91-101.

[2] Bartlett, M.S. (1973). Properties of suffciency and statistical tests. Proc. Roy.Stat. Soc. Ser. A., 160, 268–282.

[3] Bartlett, M.S. and Kendall, D.G. (1946). The Statistical Analysis of Variances Heterogeneity and the Logarithmic Transformation, J. Roy. Statist. Soc., 8, 128-138.

[4] Boos, D. and Brownie, C. (2004). comparing variances and other measures of dispersion. Stat. Sci., 19(4), 571-578.

[5] Box, G.E.P. (1953). Non-normality and tests on variances. Biometrika., 40,318-3380.

[6] Brown, B. and Forsythe, A. (1974). Robust test for equality of variances. J.Am. Stat. Assoc., 69, 364-367.

[7] Chen, Y.J., Ning, W. and Gupta A.K. (2015). Jackknife empirical likelihood for testing the equality of two variances. J. Appl. Stat. 42, 144-160.

[8] Cheng, C.H., Liu, Y., Liu, Z. and Zhou, W. (2018). Balanced augmented jackknife empirical likelihood for two sample U-statistics. Sci. China Math. 61, 1129-1138.

[9] Conover, W.J., Jahnson, M.E. and Jahnson, M.M. (1981). A Comparative study of tests for homogeneity of variances, with applications to the outer continental shelf bidding data. Technometrics. 23, 351-361.

[10] Conover, W., Guerrero- Serrano, A. and Tercero-Gomez, V. (2018). An update on “a comparative study of tests for homogeneity of variance”. J. Statist.Comput. Simulation. 88, 1454-1469.

[11] Esmailzadeh, N. (2019). A comparison of five bootstrap and non-bootstrap Levene-type tests of homogeneity of variances. Iran J. Sci. Technol. Trans. Sci., 43, 979-989.

[12] Gartsde, P.S. (1972). A Study of Methods for Comparing Several Variances. J. Amer. Statist. Assoc., 76, 342-346.

[13] Gastwirth, J.L., Gel, Y.R. and Weiwen Miao, W. (2009). The Impast of Levene’s Test of Equality of on Statistical Theory and Practice. Stat. Sci., 24,343-360.

[14] Hartlay, H.0. (1950). The Maximum F-Ratio as a Short cut Test for Heterogeneity of Variance. Biometrika., 37, 308-31.

[15] Hoeffding, W. (1948a). A class of statistics with asymptotically normal distribution Ann. Math. Statist. 19, 293-325.

[16] Hoeffding, W. (1948b). A non-parametric test for independence, Ann. Math.Statist. 19, 546-557.

[17] James, G.S. (1951). The comparison of several grups of observations when the ratios of the population variances are unknown. Biometrika. 38, 324-329.

[18] Jaffe, P.R and Parker, F.L. and Wilson, D.J. (1982). Distribution of toxic substances in rivers. J. Environ. Eng. Div. 108,639-649.

[19] Jing, B., Yuan, J. and Zhou, W. (2009). Jackknife empirical likelihood. J. Amer. Statist. Assoc., 104, 1224-1232.

[20] Layard, M. W.J. (1973). Robust Large-sample tests for homogeneity of variances. J. Amer. Statist. Assoc., 68, 195-198.

[21] Lehmann, E.L. (1951). Consistency and unbiasedness of certain nonparametric tests Ann. Math. Statist. 22, 165-179.

[22] Levene, H. (1960). Robust tests for equality of variances. In: I. Olkin, S.G. Ghurye, w. Hoeding, W.G. Madow and H.B. Mann (Eds.), Contributions to Probability and Statistice. Stanford. Univ. Press. Stanford., 278-292.

[23] Levy, K.J. (1975). An Empirical Comparison of Several Range Tests for Variances. J. Amer. Statist. Assoc., 70, 180-183.

[24] Loh, W.Y. (1973). Some modifications of Levene’s test of variance homogeneity. J. Statist. Comput. Simulation., 28, 213-226.

[25] Miller, R.J. (1968). Jackknifing variances. Ann. Math. Statist. 39, 567-582.

[26] Parra-Frutos, I. (2013). Testing homogeneity of variances whit unequal sample sizes. Comput. Stat. 28, 1269-1297.

[27] Sang, Y. (2021). A jackknife empirical likelihood approach for testing the homogeneity of K variances. Metrika. 84, 1025-1048.

[28] Sharma, D. and Kibria, G.B.M. (2013). On Some test Statistics for testing homogeneity of variances: Comparative Study. J. Statist. Comput. Simulation., 83, (10),1944-1963.

[29] Shi, X. (1984). The approximate independence of jackknife pseudo-values and the bootstrap methods. J. Wuhan Inst. Hydra-Electric Engineering. 2, 83-90.

[30] Shoemaker, L. H. (2003). Fixing the F test for equal variances, Am. Stat, 57,105-114.

[31] Wang, Q.H. and Jing, B.Y. (1999). Empirical likelihood for partial linear models with fixed designs. Statist. Probab. Lett. 41, 425-433.

[32] Wood, A.T.A., Do, K.A. and Broom, N.M. (1996). Sequential linearization of empirical likelihood constraints with application to U-statistics. J. Comput. Graph. Statist., 5, 365-385.

[33] Yitnosumarto, S. and O‘Neil, M.4E. (1975). On Levene’s test of variance homogeneity. Austral. J. Statist., 28, 230-241.