×

Comparing two independent populations using a test based on empirical likelihood and trimmed means. (English) Zbl 1470.62050

Summary: We develop a new robust empirical likelihood-based test for comparing the trimmed means of two independent populations. The simulation results indicate that the test has asymptotically correct level under various data distributions and controls the Type I error adequately for medium-size samples. For nonnormal data distributions, the power of the test is comparable to robust alternatives like Yuen’s test for the trimmed means and considerably exceeds that of the tests based on the means. In small sample settings the test version for the difference of 10% trimmed means exhibits robustness to the combined presence of nonnormality and heterogeneity.

MSC:

62G05 Nonparametric estimation
62G35 Nonparametric robustness
62G20 Asymptotic properties of nonparametric inference

Software:

EL; robustbase; WRS2
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Arnold, BC; Balakrishnan, N.; Nagaraja, HN, A First Course in Order Statistics, Classics in Applied Mathematics (2008), Philadelphia: SIAM, Philadelphia · Zbl 1172.62017
[2] Bradley, JV, Robustness?, Br. J. Math. Stat. Psychol., 31, 144-152 (1978)
[3] Casella, G.; Berger, RL, Statistical Inference (1990), Duxbury Advanced Series: Brooks/Cole Publishing Company, Duxbury Advanced Series · Zbl 0699.62001
[4] Cers, E.; Valeinis, J., EL: Two-sample Empirical Likelihood, R package version, 1, 0 (2011)
[5] Cressie, NA; Whitford, HJ, How to use the two sample t-test, Biom. J., 28, 2, 131-148 (1986)
[6] Fried, R.; Dehling, H., Robust nonparametric tests for the two-sample location problem, Stat.Methods Appl., 20, 2, 409-422 (2011) · Zbl 1337.62093
[7] Keselman, HJ; Wilcox, RR; Othman, AR; Fradette, K., Trimming, transforming statistics, and bootstrapping: Circumventing the biasing effects of heterescedasticity and nonnormality, J.Mod. Appl. Stat.Methods, 1, 2, 288-309 (2002)
[8] Lix, LM; Keselman, HJ, To trim or not to trim: Tests of location equality under heteroscedasticity and nonnormality, Educ. Psychol. Meas., 58, 3, 409-429 (1998)
[9] Marazzi, A., Bootstrap tests for robust means of asymmetric distributions with unequal shapes, Comput. Stat. Data Anal., 49, 503-528 (2002) · Zbl 1101.62321
[10] Maronna, RA; Martin, RD; Yohai, VJ; Salibián-Barrera, M., Robust Statistics: Theory and Methods (with R) (2018), New York: Wiley, New York · Zbl 1409.62009
[11] Owen, AB, Empirical likelihood ratio confidence intervals for a single functional, Biometrika, 75, 2, 237-249 (1988) · Zbl 0641.62032
[12] Owen, AB, Empirical Likelihood (2001), New York: Chapman & Hall/CRC Press, New York
[13] Qin, G.; Tsao, M., Empirical likelihood ratio confidence interval for the trimmed mean, Commun. Stat., Theory Methods, 31, 12, 2197-2208 (2002) · Zbl 1075.62554
[14] Qin, J.; Lawless, J., Empirical likelihood and general estimating equations, Ann. Stat., 22, 300-325 (1994) · Zbl 0799.62049
[15] Qin, YS; Zhao, LC, Empirical likelihood ratio confidence intervals for various differences of two populations, Syst, Sci. Math. Sci., 13, 1, 23-30 (2000) · Zbl 0973.62038
[16] Stigler, SM, The asymptotic distribution of the trimmed mean, Ann. Stat., 1, 3, 472-477 (1973) · Zbl 0261.62016
[17] Valeinis, J.; Cers, E.; Cielens, J., Two-sample problems in statistical data modelling, Math. Model. Anal., 15, 137-151 (2010) · Zbl 1191.62091
[18] Velina, M.; Valeinis, J.; Greco, L.; Luta, G., Empirical likelihood-basedANOVA for trimmed means, Int. J. Environ. Res. Public Health, 13, 10, 1-13 (2016)
[19] M. Velina, J. Valeinis, and G. Luta, Likelihood-based inference for the difference of two location parameters using smoothed M-estimators, J. Stat. Theory Pract., 13:34, 2019, doi:10.1007/s42519-019-0037-8. · Zbl 1426.62139
[20] Welch, BL, The generalization of “Student’s” problem when several different population variances are involved, Biometrika, 34, 1, 28-35 (1947) · Zbl 0029.40802
[21] Wilcox, RR, Introduction to Robust Estimation and Hypothesis Testing (2012), Boston: Academic Press, Boston · Zbl 1270.62051
[22] Wilks, SS, The large-sample distribution of the likelihood ratio for testing composite hypotheses, Ann. Math. Stat., 9, 60-62 (1938) · Zbl 0018.32003
[23] Yuen, KK, The two-sample trimmed t for unequal population variances, Biometrika, 61, 1, 165-170 (1974) · Zbl 0277.62009
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.