## A data-adaptive methodology for finding an optimal weighted generalized Mann-Whitney-Wilcoxon statistic.(English)Zbl 1162.62346

Summary: J. Xie and C. Priebe [Generalizing the Mann-Whitney-Wilcoxon Statistic. J. Nonparametric Stat. 12, No. 5, 661–682 (2002; Zbl 1058.62529)] introduced the class of weighted generalized Mann-Whitney-Wilcoxon (WGMWW) statistics which contained as special cases the classical Mann-Whitney test statistic and many other nonparametric distribution-free test statistics commonly used for the two-sample testing problem. The two-sample test that they proposed was based on any statistic within the class of WGMWW statistics optimal in the Pitman asymptotic efficacy (PAE) sense. In this paper, among other things, we show via simulation studies that for finite samples the PAE-optimal WGMWW test has substantially higher empirical power compared to the classical Mann-Whitney test for various underlying densities (especially for those densities for which Mann-Whitney test is considered a better alternative to parametric tests such as $$t$$-tests). The PAE-optimal WGMWW test is not a candidate for the practitioner’s toolbox since the corresponding test statistic contains parameters which are functions of the underlying null distribution function of the samples.
The main thrust of this paper is in introducing a data-adaptive alternative to the PAE-optimal WGMWW test, which has efficacy and power as good as the latter. We provide an estimate $$\hat{\psi}$$ for the PAE function $$\psi$$ of a WGMWW statistic, and our test is based on a $$\hat{\psi}$$-optimal WGMWW statistic. We prove strong consistency of $$\psi$$, thereby showing that our test has approximately the same efficacy as the $$\psi$$-optimal WGMWW test for large sample sizes. Via simulation studies we show that for finite samples the empirical power of $$\hat{\psi}$$-optimal WGMWW test is almost the same as $$\psi$$-optimal WGMWW test for various underlying densities. We also analyze magnetic imaging data related to subjects with and without Alzheimer’s disease to illustrate our methodology. In summary, we present a strong competitor for the classical Mann-Whitney-Wilcoxon test and many other existing nonparametric distribution-free tests, especially for moderate and large samples.

### MSC:

 62G10 Nonparametric hypothesis testing 62G20 Asymptotic properties of nonparametric inference 65C60 Computational problems in statistics (MSC2010) 92C55 Biomedical imaging and signal processing

Zbl 1058.62529
Full Text:

### References:

 [1] Adams, B.E., Ahmad, I.A. (1996). A class of mann – whitney – wilcoxon type statistics, amer. statist. 50, 324-327: comment by Adams, Adams, chang, etzel, kuo, montemayor, and schucany; and reply, Amer. statist., 54, 160, (2000) [2] Ahmad, I.A., A class of mann – whitney – wilcoxon type statistics, Amer. statist., 50, 324-327, (1996) [3] Deshpande, J.V.; Kochar, S.C., Some competitors of tests based on powers of ranks for the two-sample problem, Sankhya ser. B, 42, 236-241, (1980) · Zbl 0487.62037 [4] Grenander, U., Some direct estimates of the mode, Ann. math. statist., 36, 1, 131-138, (1965) · Zbl 0131.17702 [5] Hall, P., Limit theorems for estimators based on inverses of spacings of order statistics, Ann. probab., 10, 4, 992-1003, (1982) · Zbl 0516.60024 [6] John, M., 2005. A data-adaptive methodology for finding an optimal weighted generalized Mann-Whitney-Wilcoxon statistic. Ph.D. Dissertation, Department of Applied Mathematics and Statistics, Johns Hopkins University. [7] John, M., Priebe, C.E., 2005. A data-adaptive methodology for finding an optimal weighted generalized Mann-Whitney-Wilcoxon statistic. Technical Report No. 652, Department of Applied Mathematics and Statistics, Johns Hopkins University. · Zbl 1162.62346 [8] Jonckeere, A.R., A distribution-free k-sample test against ordered alternatives, Biometrika, 41, 1/2, 133-145, (1954) · Zbl 0058.35304 [9] Kochar, S.C., A class of distribution-free tests for the two-sample slippage problem, Comm. statist.—theory methods A, 7, 13, 1243-1252, (1978) · Zbl 0386.62034 [10] Kumar, N., A class of two-sample tests for location based on sub-sample medians, Comm. statist.—theory methods, 26, 943-951, (1997) · Zbl 0917.62038 [11] Lehmann, E.L., 1998. Nonparametrics: Statistical Methods Based on Ranks, Revised. Prentice-Hall, Englewood Cliffs, NJ. [12] Lorentz, G.G., Bernstein polynomials, (1986), Chelsea New York · Zbl 0989.41504 [13] Mann, H.B.; Whitney, D.R., On a test of whether one of two random variables is stochastically larger than the other, Ann. math. statist., 18, 50-60, (1947) · Zbl 0041.26103 [14] Marron, J.S.; Wand, M.P., Exact Mean integrated square error, Ann. statist., 20, 712-736, (1992) · Zbl 0746.62040 [15] Miller, M.I., Labelled cortical depth maps quantifying cortical change during aging, Proc. nat. acad. sci., 100, 25, 15172-15177, (2003) [16] Pitman, E.J.G., Some basic theory for statistical inference, (1979), Chapman & Hall New York [17] Priebe, C.E.; Cowen, L.J., A generalized mann – whitney – wilcoxon statistic, Comm. statist.—theory methods, 28, 12, 2871-2878, (1999) · Zbl 0944.62048 [18] Shetty, I.D.; Govindarajulu, Z., A two-sample test for location, Comm. statist.—theory methods, 17, 2389-2401, (1988) · Zbl 0642.62030 [19] Stephenson, R.W.; Ghosh, M., Two sample nonparametric tests based on subsamples, Comm. statist.—theory methods, 14, 7, 1669-1684, (1985) · Zbl 0583.62037 [20] Tryon, P.V.; Hettmansperger, T.P., A class of non-parametric tests for homogeneity against ordered alternatives, Ann. statist., 1, 6, 1061-1070, (1973) · Zbl 0275.62041 [21] Wilcoxon, F., Individual comparisons by ranking methods, Biometrics, 1, 80-83, (1945) [22] Xie, J., 1999. Generalizing the Mann-Whitney-Wilcoxon statistic. Ph.D. Dissertation, Department of Applied Mathematics and Statistics, Johns Hopkins University. · Zbl 1058.62529 [23] Xie, J.; Priebe, C.E., Generalizing the mann – whitney – wilcoxon statistic, J. nonparametric statist., 12, 661-682, (2000) · Zbl 1058.62529 [24] Xie, J.; Priebe, C.E., A weighted generalization of the mann – whitney – wilcoxon statistic, J. statist. plann. inference, 102, 441-466, (2002) · Zbl 0989.62027
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.