Abstract
A strictly nonparametric bivariate test for two sample location problem is proposed. The proposed test is easy to apply and does not require the stringent condition of affine-symmetry or elliptical symmetry which is required by some of the major tests available for the same problem. The power function of the proposed test is calculated. The asymptotic distribution of the proposed test statistic is found to be normal. The power of proposed test is compared with some of the well-known tests under various distributions using Monte Carlo simulation technique. The power study shows that the proposed test statistic performs better than most of the test statistics for almost all the distributions considered here. As soon as the underlying population structure deviates from normality, the ability of the proposed test statistic to detect the smallest shift in location increases as compared to its competitors. The application of the test is shown by using a data set.
Similar content being viewed by others
References
Baringhaus L, Franz C (2004) On a new multivariate two-sample test. J Multivar Anal 88: 190–206
Bennett BM (1962) On multivariate sign tests. J R Stat Soc Ser B 24: 159–161
Blumen I (1958) A new bivariate sign test. J Am Stat Assoc 53: 448–456
Brown BM, Hettmansperger TP (1987) Affine-invariant rank methods in the bivariate location model. J R Stat Soc B 49: 301–310
Chatterjee SK, Sen PK (1964) Non-parametric tests for the bivariate two sample location problem. Bull Calcutta Stat Assoc 13: 18–58
Dietz JE (1982) Bivariate nonparametric tests for the one-sample location problem. J Am Stat Assoc 77(377): 163–169
Good P (2005) Resampling methods: a practical guide to data analysis, 3rd edn. Birkhäuser, Boston
Govindarajulu Z, Gupta AK (1977) Certain nonparametric classification rules and their asymptotic efficiencies. Can J Stat 5(2): 167–178
Hettmansperger TP, Oja H (1994) Affine invariant multivariate multisample sign tests. J R Stat Soc Ser B 56: 235–249
Hodges JL (1955) A bivariate sign test. Ann Math Stat 26(3): 523–527
LaRocque D, Tardif S, Eeden C (2003) An affine-invariant generalization of the Wilcoxon signed-rank test for the bivariate location problem. Aust N Z J Stat 45: 153–165
Lee AJ (1990) U-Statistics theory and practice. Marcel Dekker, New York
Mann HB, Whitney DR (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 18: 50–60
Mardia KV (1967) A non parametric test for the bivariate location problem. J Roy Stat Soc Ser B 29: 320–342
Oja H, Nyblom J (1989) Bivariate sign tests. J Am Stat Assoc 84(405): 249–259
Peters D, Randles RH (1990) A multivariate signed rank test for the one sample location problem. J Am Stat Assoc 85: 552–557
Peters D, Randles RH (1991) A bivariate signed rank test for two sample location problem. J R Stat Soc Ser B 53: 493–504
Puri ML, Sen PK (1971) Nonparametric methods in multivariate analysis. Wiley, New York
Randles RH (1989) A distribution-free multivariate sign test based on inter-direction. J Am Stat Assoc 84: 1045–1050
Randles RH, Peters D (1990) Multivariate rank tests for the two sample location problem. Commun Stat Theory Methods 19(11): 4225–4238
Rempala G, Gupta AK (1999) Weak limits of U-statistics of infinite order. Random Oper Stochastic Equ 7(1): 39–52
Sen K, Mathur SK (1997) A bivariate signed rank test for two sample location problem. Commun Stat Theory Methods 26(12): 3031–3050
Sen K, Mathur SK (2000) A test for bivariate two sample location problem. Commun Stat Theory Methods 29(2): 417–436
Shetty ID, Govindarajulu Z (1988) A two-sample test for location. Commun Stat 17(7): 2389–2401
Um Y, Randles RH (1998) Nonparametric tests for the multivariate multisample location problem. Stat Sinica 8: 801–812
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mathur, S.K. A new nonparametric bivariate test for two sample location problem. Stat Methods Appl 18, 375–388 (2009). https://doi.org/10.1007/s10260-008-0095-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10260-008-0095-7