Abstract
In order to easily compare several populations on the basis of more than one feature, multivariate coefficients of variation (MCV) may be used as they allow to summarize relative dispersion in a single index. However, up to date, no test of equality of one or more MCVs has been developed in the literature. In this paper, several classical and robust Wald-type tests are proposed and studied. The asymptotic distributions of the test statistics are derived under elliptical symmetry, and the asymptotic efficiency of the robust versions is compared to the classical tests. Robustness of the proposed procedures is examined through partial and joint influence functions of the test statistic, as well as by means of power and level influence functions. A simulation study compares the performance of the classical and robust tests under uncontaminated and contaminated schemes, and the difference with the usual covariance homogeneity test is highlighted. As a by-product, these tests may also be considered in the univariate context where they yield procedures that are both robust and easy-to-use. They provide an interesting alternative to the numerous parametric tests existing in the literature, which are, in most cases, unreliable in presence of outliers. The methods are illustrated on a real data set.
Similar content being viewed by others
Notes
At distribution F, the weight function was set to \(I_{[0,q_{\delta }]}\) with \(\delta =0.025\) and \(q_{\delta }= G^{-1}(1-\delta )\) where \(G(t)={\mathbb {P}}_{F}[\mathbf {X}^t\mathbf {X}\le t]\).
References
Aerts S, Haesbroeck G, Ruwet C (2015) Multivariate coefficients of variation: comparison and influence functions. J Multivar Anal 142:183–198
Aerts S, Haesbroeck G, Ruwet C (2016) Distribution under elliptical symmetry of a distance-based multivariate coefficient of variation. Stat Pap. doi:10.1007/s00362-016-0777-4
Albert A, Zhang L (2010) A novel definition of the multivariate coefficient of variation. Biom J 52:667–675
Bartlett MS (1937) Properties of sufficiency and statistical tests. J R Soc A Math 160:268–282
Basu A, Mandal A, Martin A, Pardo L (2015) Generalized Wald-type tests based on minimum density power divergence estimators. Statistics 50:1–26
Bennett BM (1976) On an approximate test for homogeneity of coefficients of variation. In: Ziegler WJ (ed) Contribution to applied statistics. Birkhauser Verlag, Basel and Stuttgart, pp 169–171
Cator EA, Lopuhäa HP (2010) Asymptotic expansion of the minimum covariance determinant estimators. J Multivar Anal 101:2372–2388
Croux C, Haesbroeck G (1999) Influence function and efficiency of the minimum covariance determinant estimator. J Multivar Anal 71:161–190
Falk M (1997) Asymptotic independance of median and MAD. Stat Probab Lett 34:341–345
Feltz CJ, Miller GE (1996) An asymptotic test for the equality of coefficients of variation. Stat Med 15:647–658
Forkman J (2009) Estimator and tests for common coefficients of variation in normal distributions. Commun Stat Theory 38:233–251
Garcia-Perez A (2012) A linear approximation to the power function of a test. Metrika 75:855–875
Ghosh A, Mandal A, Martin N, Pardo L (2016) Influence analysis of robust Wald-type tests. J Multivar Anal 147:102–126
Ghosh A, Basu A (2013) Robust estimation for independent non-homogeneous observations using density power divergence with applications to linear regression. Electron J Stat 7:2420–2456
Gupta CR, Ma S (1996) Testing the equality of coefficients of variation in \(K\) normal populations. Commun Stat Theory 25:115–132
Hallin M, Paindaveine D (2009) Optimal tests for homogeneity of covariance, scale, and shape. J Multivar Anal 100:422–444
Hampel FR, Ronchetti EM, Rousseeuw PJ, Stahel WA (1986) Robust statistics: the approach based on influence functions. Wiley, New York
Healy MJ (1979) Outliers in clinical chemistry quality control schemes. Clin Chem 25:675–677
Heritier S, Ronchetti EM (1994) Robust bounded-influence tests in general parametric models. JASA 89:897–904
Krishnamoorthy K, Yu J (2004) Modified Nel and van der Merwe Test for the Multivariate Behrens-Fisher Problem. Stat Probab Lett 66:161–169
Krishnamoorthy K, Lee M (2014) Improved tests for the equality of normal coefficients of variation. Comput Stat 29:215–232
Ledoit O, Wolf M (2008) Robust performance hypothesis testing with the Sharpe ratio. J Empir Financ 15:850–859
Loh W-Y (1984) Bounds on AREs for restricted classes of distributions defined via tail-orderings. Ann Stat 12:685–701
Lopuhäa HP (1997) Asymptotic expansion of S estimators of location and covariance. Stat Neerl 51:220–237
Lopuhäa HP (1999) Asymptotics of reweighted estimators of multivariate location and scatter. Ann Stat 27:1638–1665
McKay A (1932) Distribution of the coefficient of variation and the extended \(t\)-distribution. J R Stat Soc 95:695–698
Nairy KS, Rao KA (2003) Tests of coefficients of variation of normal population. Commun Stat Simul 32:641–661
Pardo MC, Pardo JA (2000) Use of Rényi’s divergence to test for the equality of the coefficients of variation. J Comput Appl Math 116:93–104
Pires AM, Branco JA (2002) Partial influence functions. J Multivar Anal 83:451–468
Puri ML, Sen PK (1971) Nonparametric methods in multivariate analysis. Wiley, New York
Rousseeuw PJ, Ronchetti EM (1981) Influence curve for general statistics. J Comput Appl Math 7:161–166
Schott JR (2001) Some tests for the equality of covariance matrices. J Stat Plan Inference 94:25–36
Tsou TS (2009) A robust score test for testing several coefficients of variation with unknown underlying distributions. Commun Stat Theory 38:1350–1360
Verrill SP, Johnson RA (2007) Confidence bounds and hypothesis tests for normal distribution coefficients of variation. Research Paper FPL-RP-638. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory
Voinov VG, Nikulin MS (1996) Unbiased estimators and their applications multivariate case, vol 2. Kluwer, Dordrecht
Zhang L, Albarède S, Dumont G, Van Campenhout C, Libeer J, Albert A (2010) The multivariate coefficient of variation for comparing serum protein electrophoresis techniques in external quality assessment schemes. Accredit Qual Assur 15:351–357
Acknowledgments
The authors would like to express their thanks to Professor A. Albert (School of Public Health, University of Liege) for making the EQA data available. This work was partially supported by the IAP Research Network P7/06 of the Belgian State. They also thank the referees for providing constructive comments and helpful suggestions.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Aerts, S., Haesbroeck, G. Robust asymptotic tests for the equality of multivariate coefficients of variation. TEST 26, 163–187 (2017). https://doi.org/10.1007/s11749-016-0504-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-016-0504-4