A consistent test of independence based on a sign covariance related to Kendall's tau

The most popular ways to test for independence of two ordinal random variables are by means of Kendall's tau and Spearman's rho. However, such tests are not consistent, only having power for alternatives with ``monotonic'' association. In this paper, we introduce a natural extension of Kendall's tau, called $\tau^*$, which is non-negative and zero if and only if independence holds, thus leading to a consistent independence test. Furthermore, normalization gives a rank correlation which can be used as a measure of dependence, taking values between zero and one. A comparison with alternative measures of dependence for ordinal random variables is given, and it is shown that, in a well-defined sense, $\tau ^*$ is the simplest, similarly to Kendall's tau being the simplest of ordinal measures of monotone association. Simulation studies show our test compares well with the alternatives in terms of average $p$-values.

[1]  Kenji Fukumizu,et al.  Equivalence of distance-based and RKHS-based statistics in hypothesis testing , 2012, ArXiv.

[2]  Mario Cortina-Borja,et al.  Handbook of Parametric and Nonparametric Statistical Procedures, 5th edn , 2012 .

[3]  Kenji Fukumizu,et al.  Hypothesis testing using pairwise distances and associated kernels , 2012, ICML.

[4]  Ivy Liu,et al.  Analysis of Ordinal Categorical Data, 2nd edn by Alan Agresti , 2011 .

[5]  R. Modarres.  Measures of Dependence , 2011, International Encyclopedia of Statistical Science.

[6]  J. E. García,et al.  A non-parametric test of independence ∗ , 2011 .

[7]  C. Spearman.  The proof and measurement of association between two things. , 2015, International journal of epidemiology.

[8]  Jay Verkuilen,et al.  A Review of Analysis of Ordinal Categorical Data, Second Edition , 2010 .

[9]  Gregory E. Wilding,et al.  Empirical approximations for Hoeffding’s test of bivariate independence using two Weibull extensions , 2008 .

[10]  Maria L. Rizzo,et al.  Measuring and testing dependence by correlation of distances , 2007, 0803.4101.

[11]  Bernhard Schölkopf,et al.  Measuring Statistical Dependence with Hilbert-Schmidt Norms , 2005, ALT.

[12]  Wicher P. Bergsma,et al.  On a New Correlation Coefficient, its Orthogonal Decomoposition, and Associated Tests of Independence , 2005, math/0604627.

[13]  Maliha S. Nash,et al.  Handbook of Parametric and Nonparametric Statistical Procedures , 2001, Technometrics.

[14]  Philip H. Ramsey.  Nonparametric Statistical Methods , 1974, Technometrics.

[15]  A. Feuerverger,et al.  A Consistent Test for Bivariate Dependence , 1993 .

[16]  P. McCullagh.  Analysis of Ordinal Categorical Data , 1985 .

[17]  B. Schweizer,et al.  On Nonparametric Measures of Dependence for Random Variables , 1981 .

[18]  Paul Deheuvels,et al.  An asymptotic decomposition for multivariate distribution-free tests of independence , 1981 .

[19]  R. Forthofer,et al.  Rank Correlation Methods , 1981 .

[20]  T. Wet.  Cramér-von Mises tests for independence , 1980 .

[21]  P. Robert,et al.  A Unifying Tool for Linear Multivariate Statistical Methods: The RV‐Coefficient , 1976 .

[22]  Y. Escoufier.  LE TRAITEMENT DES VARIABLES VECTORIELLES , 1973 .

[23]  J. Kiefer,et al.  DISTRIBUTION FREE TESTS OF INDEPENDENCE BASED ON THE SAMPLE DISTRIBUTION FUNCTION , 1961 .

[24]  W. Kruskal.  Ordinal Measures of Association , 1958 .

[25]  M. Kendall.  A NEW MEASURE OF RANK CORRELATION , 1938 .