Summary
For the general two-sample problem we propose an adaptive test which is based on tests of Kolmogorov-Smirnov- and Cramèr- von Mises type. These tests are modifications of the Kolmogorov-Smirnov- and Cramèr- von Mises tests by using various weight functions in order to obtain higher power than its classical counterparts for short-tailed and right-skewed distributions. In practice, however, we generally have no information about the underlying distribution of the data. Thus, an adaptive test should be applied which takes into account the given data. The proposed adaptive test is based on Hogg’s concept, i.e., first, to classify the unknown distribution function with respect to two measures, one for skewness and one for tailweight, and second, to use an appropriate test of Kolmogorov-Smimov- and Cramèr- von Mises type for this classified type of distribution. We compare the distribution-free adaptive test with the tests of Kolmogorov-Smirnov- and Cramèr-von Mises type as well as with the Lepage test and a modification of its in the case of location and scale alternatives including the same shape and different shapes of the distributions of the X- and Y- variables. The power comparison of the tests is carried out via Monte Carlo simulation assuming short-, medium- and long-tailed distributions as well as distributions skewed to the right. It turns out that, on the whole, the adaptive test is the best one for the broad class of distributions considered.
Similar content being viewed by others
References
Büning, H. (1983). Adaptive verteilungsfreie Tests. Statistische Hefte, 24, 47–67.
Büning, H. (1991). Robuste und adaptive Tests. Berlin: Walter De Gruyter.
Büning, H. (1994). Robust and adaptive tests for the two-sample location problem. OR Spektrum, 16, 33–39.
Büning, H. (1996). Adaptive tests for the c-sample location problem — the case of two-sided alternatives. Communications in Statistics-Theory and Methods, 25, 1569–1582.
Büning, H. (2000). Kolmogorov-Smirnov and Cramèr- von Mises type two-sample tests with various weight functions. Diskussionsbeiträge des Fachbereichs Wirtschaftswissenschaft der Freien Universität Berlin, Volkswirtschaftliche Reihe,Nr11, to appear Communications in Statistics- Simulation and Computation No.4, 2001.
Büning, H. and Trenkler, G. (1994). Nichtparametrische statistische Methoden. Berlin: Walter De Gruyter.
Büning, H. and Chakraborti, S. (1999). Power comparison of several two-sample tests for general alternatives. Allgemeines Statistisches Archiv, 83, 190–210.
Büning, H. and Thadewald, Th. (2000). An adaptive two-sample location-scale test of Lepage-type for symmetric distributions. Journal of Statistical Computation and Simulation, 65, 287–310.
Burr, E.J. (1963). Distribution of the two-sample Cramér-von Mises criterion for small equal samples. Annals of Mathematical Statistics, 35, 95–101.
Canner, P.L. (1975). A simulation study of one- and two-sample Kolmogorov-Smirnov statistics with a particular weight function. Journal of the American Statistical Association, 70, 209–211.
Daniel, W.W. (1990). Applied nonparametric Statistics. Boston: PWS-Kent Publishing Company.
Fligner, M.A. and Policello II, G.E. (1981). Robust rank procedures for the Behrens-Fisher problem. Journal of the American Statistical Association, 76, 162–168.
Fueda, K. and Ohori, K. (1995). Versatile two-sample rank tests based on Wilcoxon test. Bulletin of Informatics and Cybernetics, 27, 159–164.
Garrod, J.W., Jenner, P., Keysell, G.R. and Mikhael, B.R. (1974). Oxidative metabolism of nicotine by cigarette smokers with cancer of the urinary bladder. Journal of National Cancer Institute, 52, 1421–1424.
Gastwirth, J.L. (1965). Percentile modifications of two sample rank tests. Journal of the American Statistical Association, 60, 1127–1141.
Goria, M.N. (1982). A survey of two-sample location-scale problem, asymptotic relative efficiency of some rank tests. Statistica Neerlandica, 36, 3–13.
Hàjek, J., Sidàk, Z. and Sen, P.K. (1999). Theory of rank tests. New York: Academic Press.
Hall, P. and Padmanabham, A.R. (1997). Adaptive inference for the two-sample scale problem. Technometrics, 39, 412–422.
Hill, N.J., Padmanabham, A.R. and Puri, M.L. (1988). Adaptive nonparametric procedures and applications. Applied Statistics, 37, 205–218.
Hogg, R.V. (1974). Adaptive robust procedures. A partial review and some suggestions for future applications and theory. Journal of the American Statistical, 69, 909–927.
Hogg, R.V. (1976). A new dimension to nonparametric tests. Communications in Statistics-Theory and Methods, 5, 1313–1325.
Hogg, R. V., Fisher, D.M. and Randies, R.H. (1975). A two sample adaptive distribution-free test. Journal of the American Statistical Association, 70, 656–661.
Kössler, W. (1991). Restriktive adaptive Rangtests zur Behandlung des Zweistichproben-Skalenproblems. Unveröffentlichte Dissertation, Humboldt-Universität zu Berlin.
Lepage, Y. (1971). A combination of Wilcoxon’s and Ansari-Bradley’s statistics. Biometrika, 58, 213–217.
Magel, R.C. and Wibowo, S.H. (1997). Comparing the powers of the Wald-Wolfowitz and Kolmogorov-Smirnov tests. Biomedical Journal, 39, 665–675.
Mielke, P.W.JR. (1972). Asymptotic behaviour of two-sample tests based on powers of ranks for detecting scale and location alternatives. Journal of the American Statistical Association, 67, 850–854.
Podgor, M.J. and Gastwirth, J.L. (1994). On non-parametric and generalized tests for the two-sample problem with location and scale change alternatives. Statistics in Medicine, 13, 747–758.
Randles, R.H. and Wolfe, D. A. (1979). Introduction to the theory of nonparametric statistics. New York: Wiley.
Ruberg, S.J. (1986). A continuously adaptive nonparametric two-sample test. Communications in Statistics-Theory and Methods, 15, 2899–2920.
Wilcox, R.R. (1989). Percentage points of a weighted Kolmogorov-Smirnov statistic. Communications in Statistics- Simulation and Computation, 18, 237–244.
Wilcox, R.R. (1997). Introduction to robust estimation and hypothesis testing. London: Academic Press.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Büning, H. An adaptive distribution-free test for the general two-sample problem. Computational Statistics 17, 297–313 (2002). https://doi.org/10.1007/s001800200108
Published:
Issue Date:
DOI: https://doi.org/10.1007/s001800200108